Implimenting timer_monotonic

[Gedare Bloom]

On Thu, Apr 22, 2021 at 1:39 PM zack_on_the_speed_chanel
<zack_on_the_speed_chanel at protonmail.ch> wrote:
>
> Hello,
>
> So in looking at the ticket i realized it was more the making the test and only testing the creation of a timer object. I  also have to implement the monotonic timer. Right now it seems like a monumental task for me currently.

What ticket?

We first need the test case. It can be a helpful contribution even if
you can't be the one to write the implementation, we can add the test
case code to the ticket and hopefully you or someone else can use it
when they try to make the implementation work.

> I began to  look at the code for timergettime. I'm looking at how the value of "now" (current time)  will differ  monotonic vs realtime. Currently now is set to Watchdog.ticks.  I understand in timergettime that most of it is  getting if the timer expired and putting it into the timespec.
> I also have a  question about the setting of the overrun in settimer's  _POSIX_Timer_TSR fuction. I'm wondering about how overrun is set, because it is incriminated and set to zero in the function?
>

there is pthread_kill() between increment and set to zero. So there is
a context switch here, and you don't know what might happen in
between.

>
>
>
> Thanks
> Zack
> _______________________________________________
> devel mailing list
> devel at rtems.org
> http://lists.rtems.org/mailman/listinfo/devel