In a comment on my Thoughts on DEC post, Tom Miller offered a rat’s ass if I scanned in my VAX 8600 (Venus) Ibox microcode listing.
Well, here’s Venus Ibox microcode v3.73, generated on 7-May-1984. Tom, you owe me one rat’s patootie.
This ran on Venus through the end of 1984. I had transferred to Alan Kotok‘s Simplified Architecture for Fast Execution (SAFE) project, where I was looking into VAX-11 emulation. Large VAX Engineering’s senior management called an “all hands on deck” emergency, and everyone needed do whatever they could to help Venus ship on time. So, I informally returned to work on Venus and ran one of the lab debug shifts. It was equally exhilarating and stressful… An, “I wouldn’t trade those memories for anything in the world,” and, “I hope I never go through something like that again,” experience.
This version ran VMS and oodles of VAX programs. I don’t know if there was a later version; if so, I never heard about it.
I wanted to post a picture for the rat potootie, but apparently it is not allowed. Thanks for the listing! Long long ago I ran into a bug in the BBSSI microcode that caused lots of a customer’s 8650s to hang. It took quite a while to figure out why they were hanging, but the solution was that the BBSSI instruction was being used in paged memory. A pagefault for either the target address or the location would cause the system to hang.
Why the BBSSI was in paged memory is another long story.
Due the the bug, and getting to fix it, I ended up eventually in engineering from a field software specialist job. So, I have found memories of the “buggy” BBSSI microcode.
Eh… I also wrote the Ebox microcode for the branch instructions. So that bug was probably in my code. Sorry about that!
It would be great to have the rest of the 8600 microcode in electronically readable form! Documentation for the 8600 in electronic form is, in general, kind of sparse.
I have a bit of stuff, but no more microcode. I do have an Ibox hardware printset. I’ll get them copied.