winscape 2352

Very likely.

Well ....

My understanding of what Lynn wrote is that there are several flavors of boot-IPL for the old IBM mainframe systems I know about, differing in the extent to which they preserve information about system state before the shutdown-crash. I think I'm a lot clearer on what "cold start" versus "warm start" means in the context of these systems. This fills a gap in what I remember, and adds some details, but I'm not sure it helps with the present discussion ....

Keep in mind that I probably have a better understanding (though not a perfect or complete one) of the IBM 370 architecture, even after all these years, than I have of the Intel x86 architecture(s?). I used to be fairly fluent in 370 buttembler but have never quite found time to try to learn x86 buttembler (*). I did a little work with channel programming with 370s but don't really even know what the equivalent is with an x86. And so on. So if I'm comparing to anything, it's more like to be 370 than x86.

((*) The list of stuff I really wish I knew more about seems to get longer every year. It's amazing how easy it is to put stuff on the list, whereas taking stuff off .... sometimes takes time and energy that aren't available.)

The process I'm describing here requires a running system -- either the installation program, booted from an installation medium, or a previous version-release-configuration of the operating system, booted from wherever it's normally booted from. What you're describing seems .... not quite that.

Sure -- though I think knowing a little about different systems implement the same basic processes clarifies one's understanding of the basic principles (which appeals to me in my pointy-headed academic role). I'm continuing this out in the hope that whatever basic principle I'm not getting will eventually become clear ....

-- B. L. Mbuttingill ObDisclaimer: I don't speak for my employers; they return the favor.