Father, husband, some sort of hacker, occasional computer scientist.
I love programming, fonts, computing esoterica, and learning languages (even if I’m not very good at it).
(He/him, BLM, LGBT+ and reproductive rights are human rights, left==labor, we only have one Earth.)
(Was also @lorddimwit)
| Website | https://people.well.com/~rk/ |
| Lobste.rs | https://lobste.rs/~lorddimwit |
Just made my #MonsterdonSavesChristmas donations. I'm doing my part!
I promised myself I'd get through this book this year. I probably won't, but I'm gonna get close.
It's fascinating in scope and a huge book. So many tangents to follow.
https://peterwatsonauthor.com/books/ideas-a-history-from-fire-to-freud/
In this hugely ambitious and exciting book Peter Watson tells the history of ideas from prehistory to the present day, leading to a new way of telling the history of the world. The book begins over a million years ago with a discussion of how the earliest ideas might have originated. Looking at animal behaviour […]
@rk I'm pretty sure that Xenix implemented the Sys III fork() semantics (copy code and data). I don't recall if it ported over vfork(). I don't recall segment limitations, but it could have been, at least in the original 8086 implementation. I remember SCO 80286 Xenix had an awkward limitation that the stack had to be pre-allocated, and fit in a segment. I'd think (but can't say for sure) any segment limitations should have been relieved in 80386 Xenix.
(I used to maintain the SCO Xenix FAQ.)
So. Does anyone know? How *did* they do fork on Xenix? I’m assuming they just copied entire segments around; I seem to recall Xenix requiring executables to be limited to a single data/stack segment, so forking is at most copying 128KB.
I guess I was wondering if there was something more clever going on, but that’s how I’d do it off the top of my head.
Adventures in modern AI.
I thought, hey, let me see if I can just ask AI how they implemented fork in Xenix on the 8086. That’s what all the cool kids do nowadays. Maybe I’ve been too hard on them.
This is covered in the literature. I just need a citation for further reading.
Yeah not only was it wrong about how it worked, it also straight-up lied about citations, providing fake quotes from *books that I own* to justify its incorrect description.
It’s really hard being right all the time.
Does anyone have a copy of “Porting Xenix to the Unmapped 8086” by Hare and Thomas? Proceedings of the Winter USENIX Conference 1984.
C’mon. I know one of you has to have a copy.
My HP LaserJet Pro M402n printer is so old ...
(How old is it?)
... when I use a third-party toner cartridge it whines about it being a non-HP product, but it doesn't lock up and refuse to print.
SnoopJ
G. Wozniak
Courtney Rosenthal
Mary Mazzocco
genou