Re: Before releasing Gambas 3.21.4 - Timers test failure

[Jussi Lahtinen <jussi.lahtinen@xxxxxxxxx>]

>
> I checked, and indeed the Sleep instruction repeats the system call
> until the specified time elapses.
>
> So the difference between the specified time and the actual duration of
> the instruction should come from the system.
>

And why this should not be tested?
If someone expects and relies on "Sleep 1"  sleeping for 1 second and your
implementation cannot do that, then it should be fixed, no?
Of course it is useful to give some tolerances for it.


Jussi