Experimental Physics and
Industrial Control System

Benjamin Franksen <[email protected]> · Fri, 15 Mar 2013 12:51:35 +0100

On Friday, March 15, 2013 11:58:54 you wrote:
> Hi Ben- I'm going direct to the source; if you think this would interest
> the community,
> feel free...

I think this is of general interest.

> What's the correct way to "suspend" a sequencer thread, e.g. when it
> encounters a condition where it can't go on?

There is currently no built-in feature to support that, but see below.

> In the past I've usually just printed a message and gone
> into "hibernation" by spinning in a state which just delays forever.
> The disadvantage
> there is that there's no external indication that something's wrong; it
> would be better
> if for ie. VxWorks the thread actually had a SUSPEND status so it
> triggers an alarm.
> On the other hand, I don't want to suspend the entire seq runtime, just
> that one state set.  This is for EPICS 3.14, so OSI.

I'd say you just call epicsThreadSuspendSelf from the program when you detect
an irrecoverable error. It could be you need to add a line

%%#include "epicsThread.h"

to the program (before the first state set is declared).

Would that suit your purposes?

Cheers
--
Ben Franksen
()  ascii ribbon campaign - against html e-mail
/\  www.asciiribbon.org   - against proprietary attachm€nts

Experimental Physics and Industrial Control System

Experimental Physics and
Industrial Control System