Re: [RFC] Periodic Output, Timestamped Input

[Felipe Balbi <felipe.balbi@linux.intel.com>]

[-- Attachment #1: Type: text/plain, Size: 1544 bytes --]


Hi,

Linus Walleij <linus.walleij@linaro.org> writes:
> On Tue, Dec 5, 2017 at 10:20 AM, Felipe Balbi
> <felipe.balbi@linux.intel.com> wrote:
>> [Me]:
>
>>> I was thinking about it already when implementing it but it fell
>>> out of my mind. I'm putting in on my TODO. (CC to bartosz
>>> who might be interested, he's using these ABIs quite a bit.)
>>
>> fair enough. In that case, it'll probably be easier to implement
>> HW-based timestamping with something like below:
>
> Indeed.
>
> IIO has a per-device sysfs file for setting up
> the POSIX timestamping named current_timestamp_clock.
> This was added by Gregor Boire in commit
> bc2b7dab629a
>
> The supported timestamps are:
> realtime
> monotonic
> monotonic_raw
> realtime_coarse
> monotonic_coarse
> boottime
> tai
>
> I could really need some help here... :/
>
> It feels like "monotonic" could very well be what you're
> looking for but I don't know the POSIX definition of these.
> Maybe we need to add one more for "device specific"?
>
> What I want to do is factor most of that commit out into
> lib/ and reuse for timestamping in GPIO so we can
> have coherent timestamps across GPIO and IIO.
>
> OK OK O need to get to it and stop talking...
>
> I guess it should be set per-gpiochip rather than
> per-gpio-desc though, but that makes most sense anyways.

Well, it may be that not every pin on a gpiochip supports HW-backed
timestamping. Just speculating here, that's why I made it per-gpio_desc
on the example.

-- 
balbi

[-- Attachment #2: signature.asc --]
[-- Type: application/pgp-signature, Size: 832 bytes --]