Re: [PATCH] genirq: Introduce irq_read_line()

[Bjorn Andersson]

On Tue, Oct 28, 2014 at 9:05 AM, Marc Zyngier <marc.zyngier@xxxxxxx> wrote:
> On 28/10/14 15:41, Bjorn Andersson wrote:
>> On Sat, Oct 25, 2014 at 2:22 AM, Marc Zyngier <marc.zyngier@xxxxxxx> wrote:
[..]
>> In the Qualcomm pmic we have two interrupt status registers
>> "interrupt" and "real-time". I think the "interrupt" status register
>> would be the one related to your defined constants. However what we
>> need to access in our use cases are the "real-time" status register,
>> which basically is a representation of the input to the interrupt
>> logic.
>
> Fancy. This really look like a i2c GPIO expander (not my best memories...).
>

It really is a kitchen sink, with among other things a set of gpios.

Due to limitations in address space many of the input bits are handled
through banked reads of the interrupt status bits. So that's where we
have access to gpio input, but also things like battery availability,
usb connected, charging status and so on.

>> As far as I can see the GIC does not offer anything like that, but I
>> hope we could add another constant to your enum list and utilise your
>> api for this.
>>
>> I'm not entirely sure what we should call it though,
>> IRQCHIP_STATE_LEVEL seems somewhat conflicting with level trigger and
>> the Qualcomm name IRQCHIP_STATE_REALTIME isn't very self describing.
>> Maybe IRQCHIP_STATE_LINE_LEVEL?
>
> Sure, that should be descriptive enough. I really we don't get too many
> of these though...
>
> Do you want me to wrap this into the next version?
>

Yes please.

Regards,
Bjorn
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/