Tony Harminc wrote: <begin extract> And on that note, I even more vaguely remember that SEP= (in pre-MVS OSs) was done at the channel level rather then the device, but that's so vague as to be unreliable. </end extract>
and, unlike many such, this particular vague memory is veridical. SEP= dates back to the early days of OS/360, and it was then unproblematic. Device addresses comprised of just three hexadecimal digits were part numbers, immediately decodable into their three <channel><control unit><device> components, so that, for example, 01F meant [multiplexor] channel 0, control unit 1, and device F and specified, unambiguously, the single path to/from a device. ACPs first compromised this simple scheme, and it is now all but meaningless. John Gilmore, Ashland, MA 01721 - USA ---------------------------------------------------------------------- For IBM-MAIN subscribe / signoff / archive access instructions, send email to [email protected] with the message: INFO IBM-MAIN
