Re: Multitasking question

2016-12-30 Thread scott Ford
hat, I discovered that CMS obeys its > > >> own rules. If I had unresolved WXTRN + absolute offset, it got > relocated > > >> anyway because CMS assumed the nonzero result meant the WXTRN was > > >> resolved. > > > > > >re: > > >http:

Re: Multitasking question

2016-12-27 Thread Paul Gilmartin
offset, it got relocated >> anyway because CMS assumed the nonzero result meant the WXTRN was >> resolved. > >re: >http://www.garlic.com/~lynn/2016h.html#98 A Christmassy PL/I tale >http://www.garlic.com/~lynn/2016h.html#101 Multitasking question > >part of the issue was th

Re: Multitasking question

2016-12-27 Thread Anne & Lynn Wheeler
relocated > anyway because CMS assumed the nonzero result meant the WXTRN was > resolved. re: http://www.garlic.com/~lynn/2016h.html#98 A Christmassy PL/I tale http://www.garlic.com/~lynn/2016h.html#101 Multitasking question part of the issue was that CMS had a 64kbyte os/360 system serv

Re: Multitasking question

2016-12-27 Thread Ed Jaffe
On 12/27/2016 4:08 PM, Charles Mills wrote: Very cool! Why copy? Why not use them where they are? In some cases, they are programs whose instructions are dynamically altered as needed to conform to the needs of the customer's run-time environment. In other cases, they are objects with

Re: Multitasking question

2016-12-27 Thread Charles Mills
Very cool! Why copy? Why not use them where they are? Charles -Original Message- From: IBM Mainframe Discussion List [mailto:IBM-MAIN@LISTSERV.UA.EDU] On Behalf Of Ed Jaffe Sent: Tuesday, December 27, 2016 11:01 AM To: IBM-MAIN@LISTSERV.UA.EDU Subject: Re: Multitasking question On 12

Re: Multitasking question

2016-12-27 Thread Paul Gilmartin
On Tue, 27 Dec 2016 11:01:29 -0800, Ed Jaffe wrote: > >We have literally hundreds of programs with no RLD entries i.e., no >embedded ADCONs. That makes it _really_ easy to copy them from one place >to another via MVCLE and then pass control to them without having to >worry about relocation. >

Re: Multitasking question

2016-12-27 Thread Ed Jaffe
On 12/27/2016 10:21 AM, Paul Gilmartin wrote: The address ranges wouldn't need to be identical if the code were location independent. This may have been a goal of the S/360 designers, never exploited in OS/360 software. We exploit it all the time! We have literally hundreds of programs with

Re: Multitasking question

2016-12-27 Thread Anne & Lynn Wheeler
000433f07816-dmarc-requ...@listserv.ua.edu (Paul Gilmartin) writes: > Further, with UNIX shared memory and the older CSA and LPA > a single page, possibly executable, can be mapped into multiple > address spaces. It better be reentrant. > > The address ranges wouldn't need to be identical if

Re: Multitasking question

2016-12-27 Thread Ed Jaffe
On 12/27/2016 10:24 AM, Joseph Reichman wrote: It's misleading the unit of work is really the RB Not when "unit of work" is used as a synonym for "dispatchable unit". An RB is not a unit of work. It's really just a glorified save area chain. -- Edward E Jaffe Phoenix Software International,

Re: Multitasking question

2016-12-27 Thread Joseph Reichman
It's misleading the unit of work is really the RB One more somewhat pertinent question in a address space if the unit of work is a SRB that address is ASCBSAWQ > On Dec 27, 2016, at 12:57 PM, Ed Jaffe wrote: > >> On 12/27/2016 9:51 AM, Joe Reichman wrote: >>

Re: Multitasking question

2016-12-27 Thread Paul Gilmartin
On Tue, 27 Dec 2016 09:48:59 -0800, Ed Jaffe wrote: >On 12/27/2016 9:41 AM, Joe Reichman wrote: >> The RB's associated with the task/TCB differentiate the programs but the >> code can live in same private address - range >> >> Correct ? > >Yes. That is essentially the definition of reentrant

Re: Multitasking question

2016-12-27 Thread John McKown
On Tue, Dec 27, 2016 at 11:41 AM, Joe Reichman wrote: > Hi > > Unlike other programming platforms where a program is associated with a > thread or in z/os jargon Task > > There can be number of programs running in the same private address space > under one Task aka TCB >

Re: Multitasking question

2016-12-27 Thread Ed Jaffe
On 12/27/2016 9:51 AM, Joe Reichman wrote: That term is okay the issue I have is with the term Multitasking is MultiRBing There is no such term as MultiRBing. When multitasking, each TCB represents an independent, asynchronous unit of work to the MVS dispatcher. RBs merely represent a form

Re: Multitasking question

2016-12-27 Thread Joe Reichman
: Multitasking question On 12/27/2016 9:41 AM, Joe Reichman wrote: > The RB's associated with the task/TCB differentiate the programs but > the code can live in same private address - range > > Correct ? Yes. That is essentially the definition of reentrant code. -- Edward E Jaffe Phoe

Re: Multitasking question

2016-12-27 Thread Ed Jaffe
On 12/27/2016 9:41 AM, Joe Reichman wrote: The RB's associated with the task/TCB differentiate the programs but the code can live in same private address - range Correct ? Yes. That is essentially the definition of reentrant code. -- Edward E Jaffe Phoenix Software International, Inc 831

Multitasking question

2016-12-27 Thread Joe Reichman
Hi Unlike other programming platforms where a program is associated with a thread or in z/os jargon Task There can be number of programs running in the same private address space under one Task aka TCB The RB's associated with the task/TCB differentiate the programs but the code can