On Sat, 2014-09-27 at 10:50 +0100, Jonathan Cameron wrote: > On 25/09/14 20:21, Ivan T. Ivanov wrote: > > On Thu, 2014-09-25 at 17:02 +0100, Mark Rutland wrote: > >> On Thu, Sep 25, 2014 at 10:47:15AM +0100, Ivan T. Ivanov wrote: <snip> > >> > >>>>>>> +- interrupts: > >>>>>>> + Usage: optional > >>>>>>> + Value type: <prop-encoded-array> > >>>>>>> + Definition: End of conversion interrupt number. If property is > >>>>>>> + missing driver will use polling to detect end of conversion > >>>>>>> + completion. > >>>>>> > >>>>>> Driver details shouldn't be in the binding. If the driver can poll, > >>>>>> that's good, but it should be dropped form this description. > >>>>>> > >>>>> > >>>>> Will remove driver details. > >>>>> > >>>>>> Is this the only interrupt? > >>>>>> > >>>>> > >>>>> Yes. > >>>>> > >>>>>> What do you mean be "End of conversion interrupt number"? Just describe > >>>>>> what the interrupt logically is from the PoV of the device -- interrupts > >>>>>> is a standard property so we don't need to be too explicit about the > >>>>>> type. > >>>>> > >>>>> Badly worded. Just, "End of conversion interrupt"? > >>>> > >>>> The part I didn't understand was what was meant by "End of conversion", > >>>> but dropping "number" is certainly better. > >>> > >>> It is clear now, right? End of ADC conversion. > >> > >> Sorry if I'm being thick here, but it's still somewhat confusing to me. > >> That's a consequence of me not being familiar with the HW more than > >> anything, I'm just missing simple details regarding the model of > >> operation, suchs as exactly what the "end of ADC conversion" entails. > >> There are a few things that could potentially mean depending on how the > >> HW was designed and intended to be used. > >> > >> Does the device periodically sample, convert some number of values > >> (possibly just 1), and trigger an interrupt to state that a buffer is > >> full / values are available? Or is the interrupt triggered for some > >> other reason? > > > > Interrupt is triggered after ADC convert analog signal to digital. > > Other details are irrelevant, I believe. > Often called a data ready interrupt. However, here it is per channel > so perhaps that description is confusing as well... Not exactly. There is only one interrupt. Simultaneous conversions are not possible. Regards, Ivan -- To unsubscribe from this list: send the line "unsubscribe linux-arm-msm" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html
