Add support for splatted function pointers by teor2345 · Pull Request #159643 · rust-lang/rust

teor2345 · 2026-07-21T08:42:42Z

Tracking issue: #153629

This PR adds support for splatted function pointers.

Currently splatted function pointers ICE due to unpopulated side-tables. This PR fixes the ICE by populating the side-table correctly, and fixes the MIR lowering code. It also refactors the surrounding code to reduce code duplication.

Generic function pointers also work. I'm sure if there's something extra we need to do to support all generic function pointers?

Close #158603

teor2345 · 2026-07-21T08:43:58Z

@rustbot label +C-bug

rustbot · 2026-07-23T04:23:58Z

r? @petrochenkov

rustbot has assigned @petrochenkov.
They will have a look at your PR within the next two weeks and either review your PR or reassign to another reviewer.

Use r? to explicitly pick a reviewer

Why was this reviewer chosen?

The reviewer was selected based on:

Owners of files modified in this PR: compiler
compiler expanded to 74 candidates
Random selection from 20 candidates

petrochenkov · 2026-07-23T12:16:36Z

@rustbot reroll

Inline splatted_callee() Add tests for generic function pointers

folkertdev

This PR got quite big and seemingly fixes a bunch of different things, is there any chance you can split it up? At least the refactoring into using the custom enums seems like it could be split out, just stubbing out the branches for FnPtr and errors. If possible a separate PR would be nice (just assign me as a reviewer to preserve context), otherwise separate commits would also work.

It's just really tough to review like this.

View changes since this review

folkertdev · 2026-07-25T12:15:39Z

+
+fn main() {
+    // FIXME(rustfmt): the attribute gets deleted by rustfmt
+    #[rustfmt::skip]


Might be better to put the #[rustfmt::skip] on the function to reduce noise?

Happy to do that. It's better for readability, but slightly worse for maintenance because now every line is unformatted, not just the splatted type lines.

Ultimately we'll fix this in rustfmt if function argument attributes become a thing.

folkertdev · 2026-07-25T12:17:46Z

We usually just use assert_eq!, is there any particular reason to deviate from that? Your generic<T: Tuple + Debug>(#[splat] a: T) could use format! and return a String?

No particular reason, it will be much more readable to use assert_eq! across all the splat UI tests. The generic case can just return the argument.

A codegen bug could swap the tupled argument and return value, or overwrite the return value, but the multi-argument case will catch that. Or it might be worth having a test that swaps the order of the arguments in the return value. I'll have a think about it.

folkertdev · 2026-07-25T12:25:04Z

-        receiver: Option<&'tcx hir::Expr<'tcx>>,
+        has_receiver: bool,
+        // This is the method receiver, or the function path
+        receiver_or_func: &'tcx hir::Expr<'tcx>,


this still feels kind of wrong, although I don't have any obviously better suggestions.

To make the states clearer, it could be:

enum SplattedFunc { FnDefReceiver(Expr), NoFnDefReceiver, FnPtrExpression(Expr), }

rustbot · 2026-07-25T12:30:55Z

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

teor2345

This PR got quite big and seemingly fixes a bunch of different things, is there any chance you can split it up? At least the refactoring into using the custom enums seems like it could be split out, just stubbing out the branches for FnPtr and errors. If possible a separate PR would be nice (just assign me as a reviewer to preserve context), otherwise separate commits would also work.

That's good feedback. I ended up squashing the whole thing to make it easier to rebase and rewrite, but I didn't split it into commits again.

I think it can be split into something like:

inline the splatted_callee function
refactor using custom enums (with stubs)
support for splatted FnPtrs
update tests to use assert_eq!

If you'd like some of those commits in different PRs, or in a different order, please let me know.

View changes since this review

teor2345 · 2026-07-27T07:57:09Z

-        receiver: Option<&'tcx hir::Expr<'tcx>>,
+        has_receiver: bool,
+        // This is the method receiver, or the function path
+        receiver_or_func: &'tcx hir::Expr<'tcx>,


To make the states clearer, it could be:

enum SplattedFunc { FnDefReceiver(Expr), NoFnDefReceiver, FnPtrExpression(Expr), }

teor2345 · 2026-07-27T08:00:07Z

+
+fn main() {
+    // FIXME(rustfmt): the attribute gets deleted by rustfmt
+    #[rustfmt::skip]


Happy to do that. It's better for readability, but slightly worse for maintenance because now every line is unformatted, not just the splatted type lines.

Ultimately we'll fix this in rustfmt if function argument attributes become a thing.

teor2345 · 2026-07-27T08:04:45Z

No particular reason, it will be much more readable to use assert_eq! across all the splat UI tests. The generic case can just return the argument.

A codegen bug could swap the tupled argument and return value, or overwrite the return value, but the multi-argument case will catch that. Or it might be worth having a test that swaps the order of the arguments in the return value. I'll have a think about it.

rustbot added S-waiting-on-author Status: This is awaiting some action (such as code changes or more information) from the author. T-compiler Relevant to the compiler team, which will review and decide on the PR/issue. labels Jul 21, 2026

This comment was marked as resolved.

Sign in to view

rustbot added F-splat `#![feature(splat)]` https://github.com/rust-lang/rust/issues/153629 I-ICE Issue: The compiler panicked, giving an Internal Compilation Error (ICE) ❄️ labels Jul 21, 2026

rustbot added the C-bug Category: This is a bug. label Jul 21, 2026

teor2345 force-pushed the splat-fn-ptr branch from 0037d67 to a73a1d8 Compare July 23, 2026 01:11

This comment has been minimized.

Sign in to view

teor2345 force-pushed the splat-fn-ptr branch from e91cea9 to 7c5ca3e Compare July 23, 2026 02:46

teor2345 marked this pull request as ready for review July 23, 2026 04:23

rustbot added the S-waiting-on-review Status: Awaiting review from the assignee but also interested parties. label Jul 23, 2026

rustbot assigned petrochenkov Jul 23, 2026

rustbot removed the S-waiting-on-author Status: This is awaiting some action (such as code changes or more information) from the author. label Jul 23, 2026

rustbot assigned folkertdev and unassigned petrochenkov Jul 23, 2026

folkertdev reviewed Jul 23, 2026

View reviewed changes

Comment thread compiler/rustc_hir_typeck/src/fn_ctxt/checks.rs Outdated

Make splatted FnPtr calls work (rather than ICE)

8dd6e1d

Inline splatted_callee() Add tests for generic function pointers

teor2345 force-pushed the splat-fn-ptr branch from 4207ea2 to 8dd6e1d Compare July 24, 2026 02:26

folkertdev requested changes Jul 25, 2026

View reviewed changes

rustbot added S-waiting-on-author Status: This is awaiting some action (such as code changes or more information) from the author. and removed S-waiting-on-review Status: Awaiting review from the assignee but also interested parties. labels Jul 25, 2026

teor2345 commented Jul 27, 2026

View reviewed changes

Uh oh!

Uh oh!

Conversation

teor2345 commented Jul 21, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

This comment was marked as resolved.

teor2345 commented Jul 21, 2026

Uh oh!

This comment has been minimized.

rustbot commented Jul 23, 2026

Uh oh!

petrochenkov commented Jul 23, 2026

Uh oh!

Uh oh!

folkertdev 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!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

rustbot commented Jul 25, 2026

Uh oh!

teor2345 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!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

5 participants

teor2345 commented Jul 21, 2026 •

edited

Loading

folkertdev left a comment •

edited by rustbot

Loading

teor2345 left a comment •

edited by rustbot

Loading