On Wed, Aug 13, 2014 at 6:25 AM, Walt Farrell <[email protected]> wrote:
> On Wed, 13 Aug 2014 13:16:17 +0300, Binyamin Dissen < > [email protected]> wrote: > > >I allocated some storage in key9 so that one need not switch keys to > >access/update. It seems to cause the code to use slightly more CPU. > > > > I wouldn't think there would be any noticable effect on CPU time; it's all > handled in the hardware. > Just speculation on my part, but perhaps it takes more cycles to process an instruction if PSW != StorageKey. I don't know how this is implemented in the hardware, but I remember something about a control register being involved in this "key 9" access from key 8. Perhaps this CR "lookup" does not occur when the keys match. Again, just speculation. I don't know if the POPS says anything about this or not. I haven't read it and don't really have time at this moment. > > -- > Walt > > -- There is nothing more pleasant than traveling and meeting new people! Genghis Khan Maranatha! <>< John McKown ---------------------------------------------------------------------- For IBM-MAIN subscribe / signoff / archive access instructions, send email to [email protected] with the message: INFO IBM-MAIN
