<snip> The code snippets I posted show that the code sleeps rather than waiting to be signaled which I suspect may be cheaper on those platforms. </snip> I don't know about the relative cheapness, but generally sleeping (and looking again when you wake up) is easier, and waiting/posting (pausing/releasing) is more responsive.
<snip> This is quite common in the assembler code I see. A CS loop with a timer </snip> I presume you mean a CS loop that continues to loop even if the starting condition does not change, but after a while sleeps. <snip> Is "priority inversion" not an issue on z/OS. I know that swapped out address spaces go to the top of the dispatcher queue but how does it work for tasks? </snip> Of course it's an issue. In some cases, the operating system can help. In others, the (authorized) application must do it. I say "authorized" because priority inversions in z/OS are typically multi-address-space issues, because the default for tasks in an address space is that they all run at the same priority. <snip> You can call CHAP. </snip> You can call it but it won't in general do anything to help a priority inversion case, unless it's to help a problem that you introduced by previous use of CHAP. CHAP is for creating your own intra-address space priorities. Peter Relson z/OS Core Technology Design ---------------------------------------------------------------------- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN
