Introduction to
S/370 Principles of Operation


PREVIOUS
CONTENTS
NEXT


Program Execution


In terms of our introduction to POPs, much of Chapter Five, Program Execution, is of great interest.

Here are the main sections of this chapter; the sections which we will discuss are highlighted. We'll ignore the other sections, as they reference facilities which our level of MVS does not provide.


  • Instructions
  • Address Generation
  • Instruction Execution and Sequencing
  • Dual-Address-Space Control
  • DAS Authorization Mechanisms
  • PC-Number Translation
  • Sequence of Storage References
  • Serialization

  • Normally, operation of the CPU is controlled by instructions in storage that are executed sequentially, one at a time, left to right in an ascending sequence of storage addresses. A change in the sequential operation may be caused by branching, LOAD PSW, interruptions, SIGNAL PROCESSOR orders, or manual intervention.


    Instructions

    We've already alluded to most of the contents of this section in our Imaginary S/370 Design, so very little of it should be new in terms of concepts. The actual S/370 architecture is just a bit more sophisticated than we described in our Imaginary Design. I recommend you read the entire section; about all you need to skip is the brief mention of the Vector Facility which we're ignoring since it was mainly used for specialized number crunching.

    This section also indirectly introduces the notation that POPs uses when it provides the instruction definitions we will encounter in later chapters, such as R1 and R2.

    Conceptually, the only major addition in terms of instruction capabilities beyond the Imaginary Design is that of immediate operands; in the SI instruction format, it simply means that one of the operands (I2) comes from the instruction itself rather than a separate storage location.

    If you want to get a head start on reading about instructions in POPs, you could find and skim the Move Immediate (MVI) instruction, which uses an immediate operand.


    ADDRESS GENERATION

    This section is just as important as the prior section. It is all of two pages, and should be studied carefully.

    Further instruction definition notations are also introduced here, such as B, X, and so forth.


    INSTRUCTION EXECUTION AND SEQUENCING

    Another important section, this time dealing with branching. Beginning assembler programmers can safely ignore mention of the MONITOR CALL, PROGRAM CALL, and PROGRAM TRANSFER instructions.

    Once you get to the Interruptions section, read all three paragraphs and then you can skim beginning with Types of Instruction Ending through the end of the section.


    Dual-Address-Space Control

    DAS Authorization Mechanisms

    PC-Number Translation

    My advice is that beginning assembler programmers skip these sections entirely. Sadly, our level of MVS doesn't directly support any of them.


    SEQUENCE OF STORAGE REFERENCES

    Since we're restricting ourselves to only one CPU, not much of this section has meaning until we get to the I/O chapter. The Conceptual Sequence subsection is worthy of note, as are the Instruction Fetching and Storage-Key Accesses subsections and I recommend you review them as best you can. As your mental model of the S/370 architecture improves, you'll want to return to this chapter and particularly this section.


    Serialization

    This section is more advanced material, again having mostly to do with multiprocessor systems. Beginning assembly programmers can safely skip this section, although you might find that skimming it is interesting.


    PREVIOUS
    CONTENTS
    NEXT