Experimental Physics and
Industrial Control System

Benjamin Franksen <[email protected]> · Thu, 8 Feb 2018 15:26:32 +0100

On 02/08/2018 12:00 AM, Hartman, Steven M. wrote:
> 
>> On Feb 7, 2018, at 5:31 PM, Simon Reiter <[email protected]> wrote:
>>
>>>  state check {
>>>    when(pv_mem < 2) {
>>>    } state check_valid
>>>    when(pv_rate != 0) {
>>>    } state check
>>>    when(pv_rate == 0 && delay(Waittime)) {
>>>    } state check_failed
>>>  }
> 
> I would avoid the 
> 
>    when(pv_rate != 0) {
>    } state check
> 
> in the check state as that will just loop, burning CPU. Try something like this . . .
> 
> 
>  state check {
>    when(pv_mem < 2) {
>    } state check_valid
>    when(pv_rate == 0 {
>    } state check_delay
>  }
>  state check_delay {
>    when(pv_rate != 0) {
>    } state check
>    when (pv_rate == 0 && delay(Waittime)) {
>    } state check_failed
>  }
> 

Yes, exactly. The original program with the busy loop can prevent any
threads with lower or equal priority (such as CA event threads) from
executing. This may work in some circumstances, depending on the details
of how the scheduler is configured but is not something I would rely on.
Steven's alternative with two states works reliably and does not depend
on how the scheduler works.

Cheers
Ben
-- 
"Make it so they have to reboot after every typo." ― Scott Adams

Experimental Physics and Industrial Control System

Experimental Physics and
Industrial Control System