On: Linus Torvalds on Rust support in kernel

[steveklabnik]

This code is using that allocator, that's part of the issue. He wants there to be no panics, not to have the panics handled in a special way.

[davidhyde → steveklabnik]

I see, so anything that can possibly fail should return a result instead of "maybe" panicking as a side effect? So no panics at all in the language? That may adversely affect the ergonomics of the language for user space applications. Perhaps it would be better to build a completely fallible version of the standard library and use linting to enforce no panics.

[steveklabnik → davidhyde]

At the very least, no panics for allocations. Methods that return Result exist, the issue is that while they do, the panic-ing ones also exist. There are plans to address this, they just haven't been implemented yet.

[davidhyde → steveklabnik]

Thanks, interesting to hear!

[jononor → steveklabnik]

Is a deny(panics) lint-rule feasible, now or in the short term? Something that would make a project fail to compile if anything use in it could panic?

[steveklabnik → jononor]

I don't know enough about the specifics to really say. I do know there's interest.