Thompson, Steve wrote: > I rarely called other programs, but when I did, I used the basic > protocols (R0-R1, R13-R15) with STM/LM, but using halfwords. > > When calling subroutines, R15 was the entry point, R14 the return address but I forget how parameters could be passed. I suspect that we used EXTRN/ENTRY > The BAL/R did not work, because the registers were only halfword size > which meant you had no place to store the linkage information. > > R3-R6 (as I recall, I don't happen to have my Model 20 card/booklet > handy) were fixed, R3= x'1000', R4= x'2000' ... I can't remember what > happened if you stored into them either (I'm not sure if you got a > "HALT" of the SPEC type or what). > > R0 = x'0000' R1 = x'1000' R2 =X'2000' thru to R7 = x'7000'
> R1-R2 worked exactly like the PoOP says for the other S/360s when it > came to TRT and EDMK. > > My recollection is that S/360/30 didn't support EDMK and TRT > I must disagree with the idea that it was "grossly incompatable" for > several reasons: > > a) BAL code could be moved between the two environments with a few > changes (mainly, the half-word register conventions, and using preset > registers, which is where I think the R3 for base got started...). And > we used a BAL/BALR macro so that we didn't have to change our thinking > all the time. > > b) EBCDIC is EBCDIC between the two machines (no funny stuff as happened > between Burroughs' EBCDIC and IBM's) > > c) RPG moved directly from the Model 20 to any other DOS machine (H & F > specs may have needed one or two changes) > > d) Tapes had the same internal formats between the two systems. > > e) If you had a large enough model (sufficient memory and some channel > feature(s)), you could hang 2311 disks on them and then exchange those > with a S/360 using 2311s > My recollection is that the 2311 attached to the Model 20 were uniquely model 20. There where 2 models, I think model 1 and model 11, The difference was that the model 1 had 100 cylinders and a total capacity of 2.7 MB, 100 cyl*10tracks*27000, and the model 2 had 200 cylinders giving a capacity of 5.4MB. The different models, I'm sure about, the actual figures are lost in time. My first "sysprog" type program was writing a program to backup a 200 Cyl. pack on a system that had 2 drives, a Model 1 and a model 2. It taught me a lot about the use of CCWs and EXCP Ken ---------------------------------------------------------------------- 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
