On Mon 27 Aug 18:46 PDT 2018, Lina Iyer wrote:

On Mon, Aug 27 2018 at 18:26 -0600, Bjorn Andersson wrote:

quoted

On Mon 27 Aug 09:56 PDT 2018, Lina Iyer wrote:

[..]

quoted

quoted

Thanks, I will look into Hans's RFCv2. But what would help me would be
to avoid creating the IRQ for the GPIO itself (I have the latent IRQ),
if I could just return that instead in gpio_to_irq(), it might be
easier. I understand ->to_irq() is supposed to be a translate function
only, I can avoid the dance of enabling and diabling the PDC IRQ on
suspend and resume.

I did implement gpio_to_irq() like this in the PMIC gpio/mpp drivers and
we've since concluded that we need to move this to some hierarchical
interrupt controller, because people like Linus expect to be able to say

 interrupts = <&gpio_controller 1 IRQ_TYPE_EDGE_RISING>

which is something used all over the place with the TLMM driver today.

Does it have to be &gpio_controller, can it be another interrupt controller?

Say,
	interrupts-extended = <&pdc 1 IRQ_TYPE_EDGE_RISING>;

It would require that the GPIO interrupt number space of the PDC matches
the pin numbering of the TLMM, to be somewhat maintainable.

And it would still require DT-writers to know that if the implementation
of a compatible, that references a TLMM IRQ, wants to mark the IRQ wake
capable it needs to reference the PDC instead...while still having a
pinmux/pinconf setting for the TLMM.

And for gpio_to_irq() we would need to do the mapping that you suggest,
so the TLMM still needs to have all these references to the PDC.


So I think it would be nice if we could avoid this scenario, but I don't
have any good ideas of how to do this right now...

Regards,
Bjorn