add macros for generating `Default` impls by telcharr · Pull Request #5257 · rust-lang/libc

telcharr · 2026-07-07T00:57:42Z

Description

Adds impl_default!, s_with_default!, and s_no_extra_traits_with_default! as a start for #4975. Unions get no implicit Default so a union field would need to supply one via #[custom_default(unsafe { mem::zeroed::<U>() })]. As discussed, custom_default needs to be the first attribute on a field since the macro matches it literally.

Checklist

Relevant tests in libc-test/semver have been updated
No placeholder or unstable values like *LAST or *MAX are
included (see #3131)
Tested locally (cd libc-test && cargo test --target mytarget);
especially relevant for platforms that may not be checked in CI

@rustbot label +stable-nominated

tgross35

Thank you for putting this together! It looks very workable. I have a handful of requests but all style/maintainability.

Would you be able to add a commit that makes use of this (replacing s!) in a few small places so we make sure it works in situ? Fine to do that in a separate PR, if you prefer, as long as it's coming soon.

View changes since this review

tgross35 · 2026-07-09T03:11:16Z

+/// Like [`s`], but also generates a `Default` impl for every struct in the block.
+macro_rules! s_with_default {


Nit: move this and s_no_extra_traits_with_default to directly below s_no_extra_traits! to keep them together (this file kind of flows from most important to less important / helpers)

I meant to order the macros:

s

s_paren

s_no_extra_traits

s_with_default

s_no_extra_traits_with_default

impl_default

So all the ss are together - currently impl_default is in the middle

tgross35 · 2026-07-09T04:19:54Z

+    s_with_default! {
+        pub struct S3 {
+            pub a: u32,
+            #[custom_default([0; 64])]


Could you make this something other than the default like [1; 64]? To make sure the field attribute is working rather than the derive.

I missed that there isn't currently a test for this, the change is kind of pointless without it 🙂 could you assert the field values in S3 and the other structs/unions in that area?

Actually I guess S3's pattern pattern is already checked with CustomDefault

rustbot · 2026-07-09T04:43:51Z

Reminder, once the PR becomes ready for a review, use @rustbot ready.

telcharr · 2026-07-14T03:03:24Z

Added a commit converting new/linux_uapi/linux/can.rs since it hits most cases. The only things it doesn't cover is #[cfg] and doc comments. If you'd like me to add a new commit implementing the macros that covers those cases let me know.

@rustbot ready

tgross35

A few small requests here and in the comments above, then I think this should be good to go. Could you squash please? (The change updating can.rs can stay a separate commit)

View changes since this review

tgross35 · 2026-07-20T16:18:47Z

+/// for `Default`. If it does not exist, `Default::default()` is used instead. In either case, the
+/// field is added to `processed_fields` with `#[custom_default]` stripped if necessary, and
+/// `impl_default` is invoked again with the remaining fields.
+macro_rules! impl_default {


Maybe rename this to struct_with_default or similar since it also emits the struct itself

tgross35 · 2026-07-20T16:22:52Z

+    s_with_default! {
+        pub struct S3 {
+            pub a: u32,
+            #[custom_default([0; 64])]


Actually I guess S3's pattern pattern is already checked with CustomDefault

tgross35 · 2026-07-20T16:29:25Z

+        impl ::core::default::Default for $name {
+            // Field doc comments get forwarded to the initializer alongside `#[cfg]`
+            // they're harmless there but the lint fires, so silence it.
+            #[allow(unused_doc_comments)]
+            fn default() -> Self {
+                Self { $($processed_field_defaults)* }
+            }
+        }


I'm realizing that something like #[cfg(target_arch = "x86_64")] on the whole struct may not work correctly since impl Default won't get the same config. I think we might need to filter the struct attributes and pass any cfg(...) to Default.

That can be done in a followup though, this gets us most of the way.

(Debug for unions has the same issue, guess we just haven't hit it.)

tgross35 · 2026-07-20T16:32:46Z

+        s_with_default! {
+            struct FieldAttrs {
+                #[cfg(target_arch = "x86_64")]
+                a: u8,
+                #[cfg(not(target_arch = "x86_64"))]
+                a: u64,
+            }
+        }


Could you add a #[cfg(target_arch = "x86_64")] field that doesn't have a cfg(not(...)) to go with it? Just thinking about how a: Default::default() technically works on all platforms here, but it might not always.

add default-generating macros for s! blocks

06e7c11

rustbot added S-waiting-on-review stable-nominated This PR should be considered for cherry-pick to libc's stable release branch labels Jul 7, 2026

tgross35 requested changes Jul 9, 2026

View reviewed changes

rustbot added S-waiting-on-author and removed S-waiting-on-review labels Jul 9, 2026

telcharr added 5 commits July 11, 2026 20:19

address review style/doc comments and fix cfg-gated field defaults

26a78f5

merge impl_default derives and attrs

314a599

use repetition instead of recursion in s_with_default macros

65ad3f9

wrap long lines in impl_default

b966dc7

implement default-gen macros in new/linux_uapi/linux/can.rs

3eab27d

rustbot added the O-linux label Jul 14, 2026

rustbot added S-waiting-on-review and removed S-waiting-on-author labels Jul 14, 2026

telcharr requested a review from tgross35 July 14, 2026 16:10

tgross35 requested changes Jul 20, 2026

View reviewed changes

rustbot added S-waiting-on-author and removed S-waiting-on-review labels Jul 20, 2026

		/// Like [`s`], but also generates a `Default` impl for every struct in the block.
		macro_rules! s_with_default {

Uh oh!

Conversation

telcharr commented Jul 7, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Description

Checklist

Uh oh!

tgross35 left a comment • edited by rustbot Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

tgross35 Jul 20, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Uh oh!

Uh oh!

Uh oh!

rustbot commented Jul 9, 2026

Uh oh!

telcharr commented Jul 14, 2026

Uh oh!

tgross35 left a comment • edited by rustbot Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

3 participants

telcharr commented Jul 7, 2026 •

edited

Loading

tgross35 left a comment •

edited by rustbot

Loading

tgross35 Jul 20, 2026 •

edited

Loading

tgross35 left a comment •

edited by rustbot

Loading