Tracking Issue for nested field access in offset_of

[GKFX]

Feature gate: #![feature(offset_of_nested)]

This is a tracking issue for the use of nested field access in offset_of. This is currently implemented with dot-separated fields, like in C. The original RFC for offset_of was rust-lang/rfcs#3308.

Public API

pub macro offset_of($Container:ty, $($fields:expr)+ $(,)?) { ... }

struct Struct {
    field: A,
}
struct A {
    a: ((u8, u8), bool),
}
const OFFSET: usize = offset_of!(Struct, field.a.0.1);

Steps / History

• Implementation: Included in initial offset_of implementation
• Final comment period (FCP)¹
• Stabilization PR - Stabilize offset_of_nested #128284

Unresolved Questions

• None yet.

Footnotes

1. https://std-dev-guide.rust-lang.org/feature-lifecycle/stabilization.html ↩

[est31]

This has been split off from #106655 to get the non-nested offset_off macro stabilized sooner.

How the syntax for nested access should look like is still subject to discussion, especially also how it relates to enums.

Some links:

• zulip thread about offset_of syntax
• tracking issue for enums in offset_of. Contains some of the syntax discussion.
• The PR that added enum support to offset_of also contains some syntax discussion.

[tgross35]

Could the top example be updated to show what Struct looks like?

[GKFX]

How close is this to stabilization? It looks like this is in use by the Linux kernel (https://lore.kernel.org/rust-for-linux/CAH5fLgjP98pS1wsP=YXP5Yr79Y62VF7XPKjbj+G75B3SOFt80g@mail.gmail.com/), as well as the standard library (for an enum):

rust/library/core/src/option.rs

Line 787 in 13e2d72

(self as *const Self).byte_add(core::mem::offset_of!(Self, Some.0)).cast(),

Stabilizing this may restrict future changes to the offset_of parsing logic (compiler/rustc_parse/src/parser/expr.rs:1191, parse_floating_field_access). Currently this has enum variants treated the same as fields and (with parsing support only) array indexing using [expr] syntax.

[dingxiangfei2009]

Cross-posting stabilization report and request because the two feature gates need to stabilize together for best utility.

Report

cc @Darksonn @ojeda

[Amanieu]

@rfcbot fcp merge

[rfcbot]

Team member @Amanieu has proposed to merge this. The next step is review by the rest of the tagged team members:

• @Amanieu
• @BurntSushi
• @dtolnay
• @joshtriplett
• @m-ou-se
• @nikomatsakis
• @pnkfelix
• @scottmcm
• @tmandry

No concerns currently listed.

Once a majority of reviewers approve (and at most 2 approvals are outstanding), this will enter its final comment period. If you spot a major issue that hasn't been raised at any point in this process, please speak up!

cc @rust-lang/lang-advisors: FCP proposed for lang, please feel free to register concerns.
See this document for info about what commands tagged team members can give me.