Re: [PATCH 4/4] sc16is7xx: Use threaded IRQ

[Maarten Brock <m.brock@vanmierlo.com>]

On 2020-05-18 18:57, Daniel Mack wrote:
> Hi Maarten,
> 
> On 5/18/20 1:14 PM, Maarten Brock wrote:
>> On 2020-05-17 22:44, Daniel Mack wrote:
> 
>> Summerizing:
>> - After switching to a threaded IRQ, the trigger could be switched to
>> IRQF_TRIGGER_LOW and with that interrupt sharing can be enabled for
>> this device with IRQF_SHARED.
> 
> Yes, but we don't need that. As discussed, the UART driver can cope 
> with
> edge IRQs just fine.
> 
>> - Some (your) interrupt controllers do not support IRQF_TRIGGER_LOW.
>> For those only IRQF_TRIGGER_FALLING can be used for this device and
>> thus IRQF_SHARED cannot be used.
> 
> True. Interrupts cannot be shared for this device then. That's a fair
> limitation, and it has always been like that.

It has always been like that for this driver. But that should be no
reason why the driver might not be improved. I wonder how the 8250
handles this. PC's have always shared interrupts for COM1/2/3/4 AFAIK.

>> - The driver for your interrupt controller should be improved to 
>> support
>> level IRQs.
> 
> It's a controller that sits behind another hardware bus itself, so
> polling is expensive. If the controller would need to check for level
> IRQs it would need to poll, and then we could as well just poll the 
> UART
> directly, that's just as good :)

That depends on the IRQ coming out of the interrupt controller. If that 
is
a level interrupt itself, then it is easy to see if all interrupts are
handled. Further polling zooms in on the devices that require attention.

> But again - the UART driver works perfectly fine with edge IRQs as long
> as the interrupt is not shared.

If you would require multiple sc16is7xx devices on I2C would you like to
connect multiple interrupt lines? Or just SCL,SDA and *one* IRQ?

OTOH for SPI you would require multiple CS already.

>> This makes me wonder if it would be better to let the device tree 
>> specify
>> the interrupt configuration.
> 
> There can be flags in the 2nd cell of the node, but their meaning is
> specific to the controller. Hence the SPI/I2C layers don't pass that
> information up.
> 
> What many drivers do is try with one setting, and if that fails because
> the interrupt controller returns an error, they fall back to something
> else. We could do the same here of course, but it'd be another patch on
> top, as it's unrelated to the concrete change the patch we're 
> commenting
> on is bringing in.
> 
> So what I can add is logic that first tries with IRQF_LOW|IRQF_SHARED,
> and if that fails, we fall back to IRQF_FALLING and retry. WDYT?

That sounds like a decent plan.

> 
> Thanks,
> Daniel

Kind regards,
Maarten