Brian Anderson
Brian Anderson
> Why? Consistency is important. If you know where one Rust tool puts its stuff you know where they all do.
@ollie27 I think it is yes. Like with .cargo, .git etc sometimes it's convenient to poke at internals. With rustup it seems to be important to know where the toolchains...
@mrhota Since it impacts so much of the ecosystem moving to platform-specific locations is probably RFC worthy, yes. It's probably important to be aware of all the activity in [this...
@flying-sheep I prefer to keep Rust's tooling consistent about this. Cargo and rustup are very closely related.
This thread is about moving away from the `multirust` name though. Please do open a different thread do discuss XDG etc. Edit: on second thought this thread is sufficiently about...
cc https://github.com/rust-lang-nursery/rustup.rs/issues/537
rustup needs to have concurrency support bolted on in general. Need to extract cargo's flocking code.
To start with we can just bolt on a flock to the entire runtime of rustup.
@matklad ooh awesome. Can you extract cargo's fancy flock code to its own crate to share with rustup?
Oh there's a tricky bit here in that rustup is reentrant. Can't just use a simple global lock.