On Monday, 08/28/2006 at 07:49 ZW3, "Shmuel Metz (Seymour J.)" <[EMAIL PROTECTED]> wrote:
> You don't consider access methods to be part of the operating system? Sure they are. They are controlling your access to the disk or tape. > Common services like DAIR and PARSE? Packaging vs. Academics. If they are just "helper" routines (a la gethostbyname() that [could] run in my address space without privilege) then Computer Science would not consider them part of the operating system proper. They may well be packaged with the operating system, but the operate outside of the pale. > For IBSYS/IBJOB and OS/360 IBM considered the entire code base to be > an operating system. Of course we did. Who was going to argue? :-) > >That would be, again classically, just BCP: the thing that holds the > >SVCs. > > Not all SVC's are in the BCP, and most of the BCP is not composed of > SVC's, at least not for MVS. Then perhaps my terminology is faulty. I was taught that the part of MVS that handles memory management, scheduling, dispatching, security, program management, address space management, system operation, I/O and so on was the BCP; that upon which all else is built, and whose services are accessed by SVC and [these days] PC. Further, while the service routines for the various SVCs may not be physically resident in the BCP, I understood the SVC handler itself was. Is there a better term for this? Alan Altmark z/VM Development IBM Endicott ---------------------------------------------------------------------- For IBM-MAIN subscribe / signoff / archive access instructions, send email to [EMAIL PROTECTED] with the message: GET IBM-MAIN INFO Search the archives at http://bama.ua.edu/archives/ibm-main.html
