Whilst I never had the luxury of working in the labs I did do a lot of work on implementing code on GCS. When SP4 came out I was working for the University of Salford who at the time, in conjunction with IBM UK, produced a package of software and hardware that allowed VM to be connected to an X.25 network.
The software consisted of service machine that was basically an X.25 switch, and which talked to X.25 via a Series/1 on the channel, and to other virtual machines via IUCV. When GCS and VTAM came out we modified the service machine to run in GCS. Initially it still talked to the Series/1 via the channel cards, but we then modified it to talk to X.25 via VTAM and NPSI. Appart from a few minor changes to WAITCB code and copying the LINEDIT code from CMS we made virtually no changes to the code to get it running on GCS. I am pretty sure we had access to the GCS source, (on fiche perhaps, my memory starts to fail at this point) and it looked very like the equivalent CMS code. What I really don't understand is why they didn't put more of the enhanced OS Support back into the original CMS. That would have been really usefull.... Dave P.S. still no VM at work. Corrupt HMC disks.... > -----Original Message----- > From: The IBM z/VM Operating System [mailto:[EMAIL PROTECTED] On > Behalf Of Alan Altmark > Sent: 03 April 2008 20:22 > To: [email protected] > Subject: Re: VTAM R.I.P. > > On Thursday, 04/03/2008 at 11:58 EDT, "Schuh, Richard" <[EMAIL PROTECTED]> > wrote: > > Yeah, it was a gutted OS/360 MFT system. MVS/XA was not even a gleam in > > its daddy's eye. > > My unmade point: GCS was cloned from CMS as it existed in VM/SP 3. > Additional code added to tighten file system security, support > cross-virtual machine multitasking and to run applications in problem > state is VM-written code. We did not import code from OS/360 MFT. > > I know this: I was there, writing the code for the GCS recovery machine, > as well as working on DEB security in the OS simulation for READ, WRITE, > POINT, GET, and PUT. > > Where the existing CMS MVS/SP (at the time) simulation was not sufficient > to meet VTAM's and NetView's requirements, the GCS versions of those > interfaces were updated with extra function. > > Alan Altmark > z/VM Development > IBM Endicott
