> -----Ursprungligt meddelande-----
> Från: IBM Mainframe Assembler List [mailto:ASSEMBLER-
> [email protected]] För Chris Craddock
> Skickat: den 6 januari 2012 17:45
> Till: [email protected]
> Ämne: Re: z/Arch design question.
>
> > My curiosity is why MVCLE sets the CC, thus forcing user code to branch
> back. Why not  just not update the PSW instruction address until all the
> data is processed? Still allow the interrupt like MVCL does, of course. I
> understand why the interrupt is necessary, especially in a single CP
> environment. Does anybody know? Is it a "millicode" thing?
>
>
> Well it *is* a millicode "thing" in so far as the instruction is
> implemented in millicode. So why does it work this way? The idea is that
> the instruction can come up for air somewhere in the middle and give the
> application the opportunity to do something else before
> continuing/resuming the move. Examples would be (perhaps) checking I/O
> completion, kicking off (another) I/O request, processing some business
> logic on the data being moved etc. Can't say I have ever come across a
> situation where that actually happens. But they are at least theoretically
> possible with MVCLE and *not* with the plain old fashioned MVCL.
> Realistically the only time that this would come into play is while doing
> very large moves on a busy system. Otherwise the MVCLE just does its thing
> and control passes to the next instruction.
> CC
>

I imagine MVCLE been used when handling linear datasets (VSAM) in memory.



Regards,
Thomas Berg
_________________________________________
Thomas Berg   Specialist   A M   SWEDBANK

Reply via email to