Keyboard Shortcuts
Likes
- H390-Vm
- Messages
Search
|
Re: PLC was RE: [h390-vm] RESLIB source
Sorry, I must disagree. I believe PL/C was developed at Cornell University,
toggle quoted message
Show quoted text
that is where the 'C' in PL/C comes from. We ran PL/C and all of the Waterloo software starting with VM/SP3 in 1984 through the 90s. /Fran Hensler at Slippery Rock University of Pennsylvania USA for 52 years mailto:RockFox@... "Yes, Virginia, there is a Slippery Rock" -----Original Message-----
From: [email protected] <[email protected]> On Behalf Of Brian McCullough Sent: Sunday, April 26, 2020 3:14 PM To: [email protected] Subject: Re: PLC was RE: [h390-vm] RESLIB source On Sun, Apr 26, 2020 at 12:18:52PM -0700, adriansutherland67 wrote: PLC is not PLS? Or ...Oh, no. PL/C, also known as PLC, is a version of PL/I created at Waterloo, similar to three or four other language variants designed for teaching. As I remember, there were Assembler, Fortran, Cobol and maybe others that came out of Waterloo. The memory may be wrong, but I think that they all came from the Universtity of Waterloo. Brian |
|
Re: PLC was RE: [h390-vm] RESLIB source
On Sun, Apr 26, 2020 at 03:37:15PM -0400, Bob Bolch wrote:
PLC was from Conway's team at Cornell. It was great for learning, becauseThank you. I was having trouble trusting my memory. I knew that Waterloo produced some of these, but I thought that their names started with "WAT." Of course, when I tell people of a certain age that my mother tounge was PL/I, they understand. Brian |
|
Re: PLC was RE: [h390-vm] RESLIB source
PLC was from Conway's team at Cornell. It was great for learning, because it attempted to correct student programming errors. On Sun, Apr 26, 2020, 3:31 PM Brian McCullough <bdmc@...> wrote: On Sun, Apr 26, 2020 at 12:18:52PM -0700, adriansutherland67 wrote: |
|
Re: PLC was RE: [h390-vm] RESLIB source
On Sun, Apr 26, 2020 at 12:18:52PM -0700, adriansutherland67 wrote:
PLC is not PLS? Or ...Oh, no. PL/C, also known as PLC, is a version of PL/I created at Waterloo, similar to three or four other language variants designed for teaching. As I remember, there were Assembler, Fortran, Cobol and maybe others that came out of Waterloo. The memory may be wrong, but I think that they all came from the Universtity of Waterloo. Brian |
|
Re: Program Interrupts (signals)
On Sun, Apr 26, 2020 at 06:25 PM, Dave Wade wrote:
Next time REXX is called we will re-enter from @@crt0 so we will re-establish any constantsYou see any global variables retain their values between calls. When I say global valuables I mean those linked to the RESLIB program, not those linked to the client program, of course. |
|
PLC was RE: [h390-vm] RESLIB source
¿ªÔÆÌåÓýFolks, There is a PLCT program on the 6-pack which appears to run PLC programs. Its fronted by an exec which seems unnecessary, but I had to disable IMPEX to get it to work¡ ? type world plc * ? WORLD: PROC OPTIONS(MAIN);?????????????????????????????????????? ??????WOR00010 ?? PUT SKIP EDIT(' Hello, VM World' )(A);?????????????????????????????? WOR00020 ?? END WORLD;?????????????????????????????????????????????????????????? WOR00030 ? Ready; T=0.01/0.01 18:36:35 plct world ? ? Hello, VM World ? ? ? IN STMT??? 3? PROGRAM RETURNS FROM MAIN PROCEDURE. ? Ready; T=0.01/0.02 18:36:39 ? Dave p.s. please start a new thread when the topic changes. ?ie create a new e-mail. Replying and changing the subject results in things getting in mess.. ? From: [email protected] <[email protected]> On Behalf Of pjfarley3
Sent: 26 April 2020 19:02 To: [email protected] Subject: Re: [h390-vm] RESLIB source ? I don¡¯t know if PLC is in any of the VM/370 packages but I¡¯m pretty sure there is a version on TK4- for MVS38J.? I believe it¡¯s also on the moshix 3390 file SYSCPK_3390.cckd here: ? ? HTH ? Peter ? From: [email protected] <[email protected]> On Behalf Of Bob Bolch ? Hey Doug, I wish I had those tapes. I used PLC in my first computer course, and the professor, Dr. Howard Elder, was on the PLC development team. In a later project oriented course, taught by the famous Dr. Fred Brooks, my team chopped off the PLC code generator and wrote a new one for a special Interdata computer. I learned a lot. ? I thought PLC was installed on one of the VM/370 4 pack or 6 pack systems at one time. ? Bob Bolch ? On Sun, Apr 26, 2020, 1:04 PM Doug Wegscheid <dwegscheid@...> wrote:
|
|
Re: Program Interrupts (signals)
On Sun, Apr 26, 2020 at 05:31 PM, Bob Bolch wrote:
All Noted. That might be a model for the cleanup Dave is talking about.I can't answer for Dave but for BREXX the easiest thing for me would be for a call that I could do to set an exit routine to be called as per your description; and another call so I could cancel it (or maybe just set it to zero which your mod would know to therefore to ignore). I don't mind if your mod only supports a single HT exit routine - and after all BREXX might be the only client! This would be a RESLIB or NUC program so its address would be stable.? As a matter of interest would the user memory have been cleared before the call? It does not matter either way so much but describes the scope of what the routine could expect. A |
|
Re: Program Interrupts (signals)
¿ªÔÆÌåÓýBob, ? Well I am only try to mediate. I really want a reliable REXX and was looking at ways to do it. I know a bit about ¡°C¡± but am not an expert. I know a bit about CP/CMS but its been 25 years since I last used one in anger. So I rely on questions and answers from every one to see how we can progress. ? It seems to me that what you are suggesting might allow the BREXX environment to be cleaned up¡ ¡ then I though, on the other hand, ?does it need cleaning? Next time REXX is called we will re-enter from @@crt0 so we will re-establish any constants.. However that issues a GETMAIN OR DMSFREE to obtain storage. It needs to be NUCLEUS ?storage, so its not released when a command ends¡.. So if HX does not clear system storage we need to clean up otherwise we will leak memory¡ .. therefore I think more investigation an testing is required¡
? ? From: [email protected] <[email protected]> On Behalf Of Bob Bolch
Sent: 26 April 2020 18:31 To: [email protected] Subject: Re: [h390-vm] Program Interrupts (signals) ? No. All of the CMS execution SVCSAVE control blocks ( one for each time a program calls another program in a chain, like an EXEC calling a User program, calling a CMS nucleus routine, etc.) have been deleted. All state information for a given SVC level was in those discarded blocks. Remember that I am talking about a Nucleus Extension environment introduced after VM/370. It is similar to the RESLIB capability in some respects. In both facilities, HX kills the running program, but a nucleus extension has the capability to be called for the purpose of cleaning up resources, like a pointer to memory which has been deleted . That might be a model for the cleanup Dave is talking about. ? Bob Bolch ? On Sun, Apr 26, 2020, 12:43 PM adriansutherland67 <adrian@...> wrote:
|
|
Re: RESLIB source
¿ªÔÆÌåÓýI don¡¯t know if PLC is in any of the VM/370 packages but I¡¯m pretty sure there is a version on TK4- for MVS38J.? I believe it¡¯s also on the moshix 3390 file SYSCPK_3390.cckd here: ? ? HTH ? Peter ? From: [email protected] <[email protected]> On Behalf Of Bob Bolch
Sent: Sunday, April 26, 2020 1:41 PM To: [email protected] Subject: Re: [h390-vm] RESLIB source ? Hey Doug, I wish I had those tapes. I used PLC in my first computer course, and the professor, Dr. Howard Elder, was on the PLC development team. In a later project oriented course, taught by the famous Dr. Fred Brooks, my team chopped off the PLC code generator and wrote a new one for a special Interdata computer. I learned a lot. ? I thought PLC was installed on one of the VM/370 4 pack or 6 pack systems at one time. ? Bob Bolch ? On Sun, Apr 26, 2020, 1:04 PM Doug Wegscheid <dwegscheid@...> wrote:
|
|
Re: RESLIB source
Hey Doug, I wish I had those tapes. I used PLC in my first computer course, and the professor, Dr. Howard Elder, was on the PLC development team. In a later project oriented course, taught by the famous Dr. Fred Brooks, my team chopped off the PLC code generator and wrote a new one for a special Interdata computer. I learned a lot. I thought PLC was installed on one of the VM/370 4 pack or 6 pack systems at one time. Bob Bolch On Sun, Apr 26, 2020, 1:04 PM Doug Wegscheid <dwegscheid@...> wrote:
|
|
Re: Program Interrupts (signals)
No. All of the CMS execution SVCSAVE control blocks ( one for each time a program calls another program in a chain, like an EXEC calling a User program, calling a CMS nucleus routine, etc.) have been deleted. All state information for a given SVC level was in those discarded blocks. Remember that I am talking about a Nucleus Extension environment introduced after VM/370. It is similar to the RESLIB capability in some respects. In both facilities, HX kills the running program, but a nucleus extension has the capability to be called for the purpose of cleaning up resources, like a pointer to memory which has been deleted . That might be a model for the cleanup Dave is talking about. Bob Bolch On Sun, Apr 26, 2020, 12:43 PM adriansutherland67 <adrian@...> wrote: On Sun, Apr 26, 2020 at 03:31 PM, Bob Bolch wrote: |
|
Re: RESLIB source
Bob: anytime anyone pops up with an old tape, I gotta ask if they have some other goodies stashed away...???? I am looking for old (pre MVS/SP) JES2 modifications tapes, and for an MVT/MVS Cornell PL/C tape (actually a VM PL/C tape would be great, too).???? Any chance you have either of those squirreled away? ?
On Sunday, April 26, 2020, 9:20:11 AM EDT, Bob Bolch <bob@...> wrote:
I found one source version on a 1980 Waterloo tape.??
Bob Bolch ---------- Forwarded message --------- From: Bob Bolch via <Bob=[email protected]> Date: Sun, Apr 26, 2020 at 8:51 AM Subject: [h390-vm] RESLIB source To: <[email protected]> Dave,
Is the Cornell RESLIB code loaded on the SIXPACK disks? I can't find it. Thanks/Bob Bolch |
|
Re: Program Interrupts (signals)
I ran some NUCXLOAD tests in VM/ESA 2.4 by specifying the SERVICE operand. The test nucleus extension does a CMSCALL to a program with a loop in it. I issued HX during the loop. The test nucleus extension?does indeed get control at it's entry point after the HX. However, the command that was HX'd?is over. The call to the nucleus extension is a new invocation?made via a CMSCALL from the Nucleus. There's no mechanism to know what the HX'd?program was doing. When the SERVICE call? entry to the Nucleus Extension returns, it returns to CMS, and we are back at the Ready prompt. Some kind of cleanup call to the DMSREX command might be done with some CMS mods and some RESLIB mods. The new call would not get any information about the ABENDing instantiation. Dave, would this allow the cleanup you want? Bob Bolch Bob Bolch On Sun, Apr 26, 2020 at 8:26 AM Bob Bolch via <Bob=[email protected]> wrote:
|
|
Re: RESLIB source
¿ªÔÆÌåÓýBob, You need to do CMSACC then its in DMSRES ASSEMBLE B Dave ? From: [email protected] <[email protected]> On Behalf Of Bob Bolch
Sent: 26 April 2020 13:51 To: [email protected] Subject: [h390-vm] RESLIB source ? Dave, ? Is the Cornell RESLIB code loaded on the SIXPACK disks? I can't find it. ? Thanks/Bob Bolch |
|
Re: Program Interrupts (signals)
On Sun, Apr 26, 2020 at 12:25 PM, Mark L. Gaubatz wrote:
there is an argument for HX to be trappable by OSI hope you don't mind me saying, in this purely academic discussion, that it feels that you are conflated; on one hand you don't want HX to be trapped by user programs, on the other you know it's a required and valid capability! :-)? |
|
RESLIB source
I found one source version on a 1980 Waterloo tape.?? Bob Bolch ---------- Forwarded message --------- From: Bob Bolch via <Bob=[email protected]> Date: Sun, Apr 26, 2020 at 8:51 AM Subject: [h390-vm] RESLIB source To: <[email protected]> Dave,
Is the Cornell RESLIB code loaded on the SIXPACK disks? I can't find it. Thanks/Bob Bolch |
|
Re: Program Interrupts (signals)
On Sun, Apr 26, 2020 at 01:04 PM, Bill Turner, wb4alm wrote:
Now my mind is a bit fuzzy, because this was many years ago, but as I recall I used "HI" frequently, but only used "HX" once.You are right (but not fuzzy), if we have HI to halt REXX it's all good and HX becomes much less of an issue. I agree with folks here ... most CMS users know / expect HX to be terminal and I seem (many years ago) to have learnt that an IPL would be sensible thing to do straight afterwards! |
Fran Hensler