I believe the only true atomic instructions are CS and CSD. PLO is considered 
atomic when using the same lock. Concurrency is only guaranteed with some sort 
of lock which is to expensive to implement for every instruction and for 
storage being referenced / modified. The odds of a single instruction 
concurrency issue is extremely low (essentially 0) but still possible. CPU's, 
storage, and pipelines are exactly the same. Using MVC example, the speed of 
movement would be exactly the same therefore you should never experience a 
problem. If on the other hand, there was something to make one MVC operate at a 
different speed (e.g. pipeline algorithm), then you would have a concurrency 
issue.
Regards, Jon.  
 On Sunday, July 30, 2017 5:50 PM, Phil Smith <[email protected]> wrote:
 
Mmm...I'm pretty sure a single instruction is still atomic. I'm sure Peter 
Relson or one of the other IBMers will chime in, but it there has to be some 
sort of interlock at some level. And I've debugged plenty of concurrency 
problems, never seen a mixture from a single instruction!

...phsiii

   

Reply via email to