Skip to content

whitelist some ARM features #53926

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
merged 1 commit into from
Sep 3, 2018
Merged
Changes from all commits
Commits
File filter

Filter by extension

Filter by extension

Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
4 changes: 4 additions & 0 deletions src/librustc_codegen_llvm/llvm_util.rs
Original file line number Diff line number Diff line change
Expand Up @@ -86,10 +86,14 @@ unsafe fn configure_llvm(sess: &Session) {
// array, leading to crashes.

const ARM_WHITELIST: &[(&str, Option<&str>)] = &[
("aclass", Some("arm_target_feature")),
Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

As for mclass and rclass too, this should be renamed to a-profile before stabilisation. IMO it's better to do that now to avoid any unstable code using it to have to be changed.

Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Also, before stabilization, neon on aarch64 should be renamed to asimd. @japaric it might be worth it to open an issue to track these on stdsimd, there might already be an issue open for the asimd rename task, so maybe this could be added to that one.

Copy link
Member Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Hmm, how does stabilization work for target_features? #[cfg(target_feature = "mclass")] is not feature gated on stable but it doesn't work. AFAICT that feature gate will work in the next beta which comes out in a week unless there's some "this is a nightly only feature" logic that I didn't see.

Copy link
Contributor

@gnzlbg gnzlbg Sep 4, 2018

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Hmm, how does stabilization work for target_features?

In the RFC for the intrinsics, you write which target features they need, and that's pretty much it. We can stabilize them with the intrinsics.

AFAICT that feature gate will work in the next beta which comes out in a week unless there's some "this is a nightly only feature" logic that I didn't see.

Which feature gate? for example, the aclass above is gated on feature(arm_target_feature). Once arm_target_feature is stabilized, the aclass target feature will be usable on stable Rust.

Before stabilization we should split arm_target_feature features that we are not stabilizing into a new unstable feature.

("mclass", Some("arm_target_feature")),
("rclass", Some("arm_target_feature")),
("dsp", Some("arm_target_feature")),
("neon", Some("arm_target_feature")),
("v5te", Some("arm_target_feature")),
("v6k", Some("arm_target_feature")),
("v6t2", Some("arm_target_feature")),
("v7", Some("arm_target_feature")),
("vfp2", Some("arm_target_feature")),
("vfp3", Some("arm_target_feature")),
Expand Down