GCC-Rust Feedback Sought - Possibly Aiming For Upstream In GCC 13

[phoronix]

Phoronix: GCC-Rust Feedback Sought - Possibly Aiming For Upstream In GCC 13

The Rust-GCC front-end that allows Rust code to be compiled with the GNU Compiler Collection (GCC) could possibly be upstreamed for next year's GCC 13 compiler release but not necessarily complete at that stage. In any case, it's good seeing progress on Rust-GCC as an alternative to Rust's official LLVM-based compiler...

https://www.phoronix.com/scan.php?page=news_item&px=GCC-Rust-Upstream-Talk-Thread

[bachchain]

This date does not account for the borrow checker feature [...], which we have a plan to implement, but this will be a further six-month project.

Great, a full six-month window for FUDers to proclaim that Rust's memory safety guarantees are meaningless.

[archkde]

Great, a full six-month window for FUDers to proclaim that Rust's memory safety guarantees are meaningless.

It's not Rust's fault if an unofficial implementation omits a very important feature.

[bug77]

Great, a full six-month window for FUDers to proclaim that Rust's memory safety guarantees are meaningless.

It's usually a good strategy to upstream what you have and wait for feedback. But Rust without the borrow checker... who's going to provide feedback on that?

[pipe13]

It's not Rust's fault if an unofficial implementation omits a very important feature.

Please read TFA. GCC is the official compiler for the Linux kernel. What other frameworks do you not wish Rust to be part of?

[RahulSundaram]

It's usually a good strategy to upstream what you have and wait for feedback. But Rust without the borrow checker... who's going to provide feedback on that?

https://github.com/Rust-GCC/gccrs/wi...orrow-checking is their strategy. The feedback asked for here is from the GCC project. They don't depend on any specific features to provide input on how best to merge the code.

[bug77]

https://github.com/Rust-GCC/gccrs/wi...orrow-checking is their strategy. The feedback asked for here is from the GCC project. They don't depend on any specific features to provide input on how best to merge the code.

Oh, I didn't mean to imply upstreaming now is useless or wrong. Just that it will probably generate below average feedback levels.

[cb88]

Oh, I didn't mean to imply upstreaming now is useless or wrong. Just that it will probably generate below average feedback levels.

Doubtful.... its almost certain all the code this would be tested against would have already had the rustc borrow checker ran against it... so its a moot point. In any case this would allow a nicer route to quick bootstraping of rust without having to compile a ton of old versions, as well as allow rust code to be built places were its harder with rust infrastructure.

Anyway that is the beauty of a compile tile borrow checker...

[ssokolow]

It's not Rust's fault if an unofficial implementation omits a very important feature.

IIRC, Rust is a trademark, just like Linux and Firefox and various other things, with the intent that the mark holder can impose requirements to ensure use of the name doesn't reflect badly on them.

[dragon321]

Please read TFA. GCC is the official compiler for the Linux kernel. What other frameworks do you not wish Rust to be part of?

For Linux maybe but for Rust not really. It doesn't matter here how important is GCC for Linux, for Rust currently reference implementation is rustc and that what defines features of this language, not unofficial GCC port. Sure, GCC port is nice thing and will be useful but it doesn't define Rust features or development.