IRC channel logs

2025-05-30.log

back to list of logs

<pabs3>sigh rust https://blog.rust-lang.org/inside-rust/2025/05/29/redesigning-the-initial-bootstrap-sequence/
<pabs3>"The standard library will move from supporting being built by both the previous toolchain version and the current version to only supporting the current version."
<lanodan>what
<lanodan>How do they even want to do this without requiring gigantic obscure blobs
<lanodan>Although that's assuming they think of this as a problem while rustc has been just that since the beginning
<lrvick>Wait so you need the current version of the rust toolchain to build the current version of the rust toolchain?
<lrvick>Do we need to build every commit?! lol
<lrvick>surely the last rustc+stdlib can build next rustc which builds next stdlib
<lanodan>Yeah hopefully rustc doesn't also needs latest stdlib otherwise well… Rust is kind of done for given software doesn't seems like it can do LTS due to the npm-like ecosystem.
<stikonas>well, it now says on the page that distribution packages will still be able to build N + 1 version of Rust using version N of Rust
<lrvick>whew.
<agg1>"Without any vendor-neutral standard followed development cannot agree upon which releases or versions of rustc could compile itself self-hosting"
<agg1>maybe there's a better phrasing to cite this comedy
<agg1>EU/UK were throwing hundreds of millions at this nonsense, as far as i could tell (helsing.ai for example placed some job vacancies for rust-lang prominently)
<agg1> http://tinyfront.mooo.com/docs.html#Acceptance-Criteria
<lrvick>Meanwhile in an opposite direction
<lrvick>I was able to build i386 llvm/clang from live-bootstrap base
<lrvick>going to stop building live-bootstrap steps at gcc10 and see if I can pivot from there
<agg1>hasn't been live-bootstrap base been gcc-4.x?
<lrvick>Well live-bootstrap currently goes all the way up to gcc-13 which I am using to build llvm/clang
<agg1>ok. with "baseline" i was thinking of something different