Time for a new #OpenBSD story! Investigating what appears to be a VAX bug turns out to be the consequence of a more-than-late processor design change, so brace yourselves for a VAX history lesson!

http://miod.online.fr/software/openbsd/stories/vaxfp.html

Math is hard

@miodvallat Man, that last bit makes me wish cpus (with variable length instructions) had two instruction pointers, current and next, or something. Three would be cooler, but I’m not sure what to make the third mean…

@guenther Yes, that would have been addressing this much easier.

As for the third pointer, that's easy: it points to the first non-bogus instruction in the program.

Writes are ignored, and upon read, its value is UNPREDICTABLE.

@miodvallat Ah, so the 'f' in 'sfip' is 'first'!
@guenther Actually, it's fetch.