Happy Birthday VAX 11/780 (influence of)

On 2010-10-28 23:02, Rich Alderson<RichA at vulcan.com> wrote: One more thing here. This is not directly to you, Rich, but this line is as good an introduction for me throwing another wrench in the machinery as any... And I do love wrenches. It has been stipulated that the PDP-11 don't have virtual memory, but the VAX do, and various arguments for this view. As far as I can tell (and people, please correct me if I've gotten this wrong) the reason for this view boils down to: 1) The physical address space is larger than the virtual on a PDP-11. 2) The PDP-11 don't load just parts of the memory and handle page faults as a way to do demand paging. Now, as for #2, I think I have established that some models of the PDP-11 could in theory do this, if you just wrote the software for it. The fact that noone have does not take away the capability. If anyone disagree with this, I'll be happy to explain exactly how the code should look like on the PDP-11 to do demand paging. As for #1, I now drag out DEC STD 032 as exhibit A. It's a wonderful document, more commonly known as the VAX Architecture Reference Manual. More specifically, the version called EL-000 32-00. Anyone can read it, it's on bitsavers. http://www.bitsavers.org/pdf/dec/vax/archSpec/EL-00032-00-decStd32_Jan90.pdf Now, I direct you all to page 4-7 of that document, which talks about physical addresses on a VAX. Notice how a physical address on a VAX can be 34 bits, while the virtual address is only 32 bits. Next question: Does the VAX not have virtual memory any more now that I've pointed this out? Or do you need to redefine virtual memory in yet a new and strange way to exclude the PDP-11... :-D Johnny -- Johnny Billquist || "I'm on a bus || on a psychedelic trip email: bqt at softjar.se || Reading murder books pdp is alive! || tryin' to stay hip" - B. Idol