Notices by John Regehr (regehr@mastodon.social), page 2

what’s missing from the index for this textbook about low-level C programming?? 🤔

ugh!

last night and today

wow I just left a comma out of an array initializer so the compiler did string concatenation and that was more confusing than I'd have expected

one of the nastiest tidbits from the Therac-25 incident -- the people testing the software weren't proficient with it and therefore had a hard time triggering the abundant latent race conditions in the interface

don't let anyone tell you that software testing isn't creative work, it is!!

https://web.stanford.edu/class/archive/cs/cs240/cs240.1236/old//sp2014/readings/therac-25.pdf

I've almost stopped assigning Therac as reading for students, it's so old that they just don't feel the connection. which is a shame! cause shitty software and hubris are timeless.

the linked document should be mandatory reading, but it's tough stuff -- contains some highly visceral subjective descriptions of being inside a radiation therapy machine when it runs amok

is there a good reason all 3 x86-64 compilers use lea instead of inc or add to increment a long?

https://gcc.godbolt.org/z/YnoW75zav

@zwarich I've still not written any non-trivial Rust. for compiler development, I want classes, GC, pattern matching, and of course memory safety. but I don't want to write OCaml :). maybe Swift is the compiler language if Rust isn't it.

so far I don't think we've seen an obviously plausible LLVM successor, but it's nice that people are trying!

https://yasserarg.com/tb

"Compiler Fuzzing in Continuous Integration: A Case Study on Dafny"

https://www.doc.ic.ac.uk/~afd/papers/2025/ICST-Industry.pdf

@lzg David Lynch is apropos right now, and he will save you some effort by taking over the task of hallucinating things

@andrewrk hand twins!!

the prettiest arrowhead I ever found in Utah!

@lzg Sarah and I like to watch Silo and every single episode we say "it's too dark" and "turn on subtitles I can't hear anything"

@dalias @Sobex @dave_andersen @va2lam @zwol @chipchristian yeah I imagine that at present, a lot of code works only due to separate compilation

@dalias @Sobex @dave_andersen @va2lam @zwol @chipchristian ah, no problem then!

@dalias @Sobex @dave_andersen @va2lam @zwol @chipchristian I thought effective type was conferred when you first store to the typeless memory, and remains the same after that

@dalias @Sobex @dave_andersen @va2lam @zwol @chipchristian ah, super cool. not something I can verify with quick inspection :).

but then eventually, memory is going to be repurposed, for example when you split a block that has already been used by the client. in that case we're just relying on separate compilation, I suppose?

@Sobex @dave_andersen @va2lam @zwol @chipchristian hey @dalias -- I just took a super quick look at MUSL's mallocng and it seems to be dereferencing typecast pointers in ways that might violate strict aliasing. do you mind telling us why this isn't a problem? maybe you insist that this is compiled with strict aliasing disabled?