Re: waittime function precision

On 4/15/2014 7:04 AM, Jevgeni Astanovski wrote:

I've used waittime function for pretty long time in cases, where I needed to have a delay in my program.
But recently a question appeared - how precise it is and are there any alternatives that allow to make delays for less than 1/100 of a second.

I think this is expected behaviour for waittime(). I always understood
waittime to guarantee a minimum delay, not to guarantee a precise delay.

_WAIT_NORMAL sets all bits off, so
Access state control = no explicit transfer of PAG
MPL control = remove thread from multiprogramming level
Async signals = Not allowed

Judging from this, it seems like there would be some time the processor
spends swapping the thread into and out of the MPL. It would be
interesting to try this on a dedicated machine, and also with
_WAIT_MPL_SET rather than _WAIT_NORMAL.

http://publib.boulder.ibm.com/infocenter/iseries/v5r4/index.jsp?topic=%2Frzatk%2FWAITTIME.htm

--buck