Experimental Physics and
Industrial Control System

Eric Norum <[email protected]> · Tue, 26 Nov 2002 14:50:25 -0600

On Tuesday, November 26, 2002, at 02:25  PM, Marty Kraimer wrote:

At the EPICS meeting last week Till asked:

1) Could we declare a maximum priority for any component of base?
2) Why does iocsh run at epicsThreadPriorityMax?

Brief discussion.

1) sounds like a good request. How about

epicsThreadPriorityBaseMax = 91

2) I will assume that this was done to be like vxWorks, which runs the

vxWorks shell at highest priority. If iocsh runs at

epicsThreadPriorityBaseMax it seems like it should be OK. If a very

high prority application thread uses all the CPU time iocsh can't get

control but this is the applications problem.

 (2) -- I think we did this to ensure that you could still get some

response at the console even if some other task went insane and started

gobbling CPU cycles. With a strict priority-based scheduler like RTEMS

and vxWorks you'd have no possiblity of dealing with such tasks if the

shell were at a lesser or equal priority.

--
Eric Norum <[email protected]>
Department of Electrical Engineering
University of Saskatchewan
Saskatoon, Canada.
Phone: (306) 966-5394   FAX:   (306) 966-5407

Experimental Physics and Industrial Control System

Experimental Physics and
Industrial Control System