Re: [Patch v3 03/11] irqchip: axi-intc: Add support for parent intc

From: Marc Zyngier
Date: Thu Sep 01 2016 - 10:48:42 EST


On 01/09/16 14:52, Zubair Lutfullah Kakakhel wrote:
> Hi,

[...]

>> But that still doesn't address the case I had in mind, which is when you
>> have *two* AXI-intc, one cascaded into the other. Is that something that
>> could be built? You should at least make sure that there is a big fat
>> warning if you don't want to support that case, because that will be
>> hell to debug.
>
> Oo. I didn't think of that one. tbh, I'm not sure if that is currently supported
> because this driver came out of arch/microblaze. And it didn't have any code that
> looked supportive of chained interrupt handling.
>
> 'Technically', it should be possible to synthesize two daisy chained axi interrupt
> controllers on an FPGA. But I don't see it being supported before.
>
> A warning would be nice. Any suggestions on the most suitable way?

Check if you've already allocated a xintc_irqc. If so, abort the probing
early...

Thanks,

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