#RUST LANGUAGE RELEASE DATE FULL#
#RUST LANGUAGE RELEASE DATE CODE#

One example here:, these bugs were quite rare in the wild, but the deprecation window lasted from December 2018 (deprecated behavior started emitting warnings) to December 2019. Even such fixes have been made gradually over multiple releases, not as flag day releases. They have happened, but very rarely and with minimal impact as far as I have seen.

Rust only breaks backwards compatibility in 1.x releases is when necessary to fix safety/soundness bugs. > If your security is gained by constant refactoring due to depracations The second is why keeping a GCC backend up to date is complicated.

#RUST LANGUAGE RELEASE DATE CODE#

Like Linux, they have a stable 'user space interface' (language definition) and internal unstable interfaces (intermediate forms passed to code generation). Since the Rust 1.0 release in May 2015, they have committed to backwards compatibility. You're presumably using a kernel with a ~10-week release cycle, would you claim that fact makes Linux unstable and unsuitable for sizable projects? The parent commenter was talking about keeping an independent compiler up to date with new Rust features, not about keeping Rust code working and relevant. The fact that they release frequently does not mean that they break compatibility often. It's unlikely to be a path forward for supporting additional architectures.

It's not designed to generate efficient code it's designed to generate functional code, so that you can bootstrap rustc without a Rust compiler (by compiling an old version of rustc with mrustc, then walking forward a few steps to current rustc). Mrustc is mostly a bootstrapping tool it doesn't compile current Rust, and it has no safety checks whatsoever, so it's only useful on code you've already tested with another compiler. Using mrustc would potentially let Rust be compiled via the same C compiler that was compiling the rest of the kernel. > The mrustc project is an experimental Rust compiler that emits C code. I'd also love to have the competition in code generation backends, to make sure we're getting the best code generation possible, and to enable cross-language LTO with either GCC or Rust. In any case, there are many paths to supporting all the architectures people are willing to support.

#RUST LANGUAGE RELEASE DATE FULL#

It'd take some more work to turn this into a full solution, but it's a potential option, and it would avoid unnecessarily reimplementing the frontend of rustc.Īnother possibility mentioned at LPC would be a rustc backend that uses libgccjit to generate code despite the name, libgccjit also works for ahead-of-time code generation. This project is in an early stage, but it presents another avenue toward closing the architecture gap in the future.Ĭontext here: I was mentioning the nascent gcc-rust project at, which compiles Rust's 'MIR' intermediate representation. In addition, Triplett cited some interest in - and work toward - a Rust front end for GCC, potentially enabling Rust to target any architecture GCC supports.

Rust language release date