Fiche was fun. Up to 370/148, microcode fiche was provided, including microcode problems.
Browsing, I found one problem described as, "When LM instruction specifies same register twice -- that is, to load one register -- all 16 registers are loaded". Oops. And once, working on an interesting but stupid project that involved validating opcodes behaving properly (because management heard loony advice from an alleged industry expert that 4341 wouldn't be fully compatible architecture) several opcodes that shouldn't have existed did, didn't 0C1. Microcode fiche revealed that they were assist instructions for VS/1, etc. On Mon, 31 Oct 2022 23:46:40 -0500, Brian Westerman <brian_wester...@syzygyinc.com> wrote: >I agree, I can't possibly convey how much I learned from some old IBM fiche >that I had access to in the computer center when I just started college. The >first really important thing I wrote (I was 17) were mods to pass the >condition codes from step to step within JES2 and then send the highest one to >the console and syslog at job end. I later learned that others had done the >same thing, and long before me, but I learned a lot. That code didn't work >with the first version of MVS I was exposed to after college, so it was >followed by doing that same thing with two jes exits and then even later >writing our companies Automation software that pulls the condition codes from >the same fields they were placed in originally way back then. > >Everything I have written over the years is still based on concepts and >techniques that I first learned by looking at the code in the IBM fiche. > >I had an extra advantage in that I worked for IBM throughout that same time >and was able to see some truly spectacular coding techniques and I am truly >thankful for that opportunity. > >I realize that IBM wanted to keep nefarious people from copying the code, but >I think that we lost a great deal of experience and expertise when we lost >access to the code. Some of those techniques are just not around for people >to examine and learn from, and that's very sad. > >Brian > > > >On Mon, 31 Oct 2022 19:42:50 -0400, David Spiegel <dspiegel...@hotmail.com> >wrote: > >>Hi Tom, >>1983, eh? >>The same year as the (expletive deleted) OCO policy. >>I've seen IBM-lifers defend it on this forum, yet, it still did not/does >>not make sense. >> >>Regards, >>David >> > >---------------------------------------------------------------------- >For IBM-MAIN subscribe / signoff / archive access instructions, >send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN ---------------------------------------------------------------------- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN
