Expose core::intrinsics::abort as std::process::abort

[apasel422]

This would be a safe wrapper around the intrinsic.

[nagisa]

Semantics of this intrinsics are significantly more different compared to C
abort, so I'm not sure whether this implementation strategy is the best one
to take.
On Nov 11, 2015 2:29 AM, "Andrew Paseltiner" [email protected]
wrote:

This would be a safe wrapper around the intrinsic.

—
Reply to this email directly or view it on GitHub
#1363.

[Stebalien]

I don't know the details of abort but I'd like some safe wrapper.

[Gankra]

Even a stable wrapper would be grand :)

[fitzgen]

This would be super helpful for wasm stuff, where panics and aborts translate into traps either way, but panics pull in all the formatting and panicking infrastructure, which adds something like 30-40K to binary size.

Anything I can do to help push this forward?

[fitzgen]

Er wait, I'm not thinking of std::process::abort, which already exists.

I'm thinking of a stable core/#![no_std] story for abort. Maybe that is a separate issue somewhere?

[nagisa]

rust-lang/rust#44489 is where you’d want to discuss panic infrastructure and it pulling in formatting machinery unnecessarily.

[Centril]

This is done since std::process::abort is stable.

[nagisa]

std::process::abort is stable, but it is not std::intrinsics::abort. It is not clear though if the original request was for a stable wrapper around std::intrinsics::abort or for anything that could be used like an abort.

[Centril]

@nagisa Weird, the documentation says: "The stabilized version of this intrinsic is std::process::abort".

https://doc.rust-lang.org/nightly/std/intrinsics/fn.abort.html

[Centril]

Closing in favor of #2512.