Enforce that `dyn*` coercions are actually pointer-sized #104338

compiler-errors · 2022-11-12T23:05:32Z

Implement a perma-unstable, rudimentary PointerSized trait to enforce dyn* casts are usize-sized for now, at least to prevent ICEs and weird codegen issues from cropping up after monomorphization since currently we enforce nothing.

This probably can/should be removed in favor of a more sophisticated trait for handling dyn* conversions when we decide on one, but I just want to get something up for discussion and experimentation for now.

r? @eholk cc @tmandry (though feel free to claim/reassign)

Fixes #102141
Fixes #102173

rustbot · 2022-11-12T23:05:39Z

Hey! It looks like you've submitted a new PR for the library teams!

If this PR contains changes to any rust-lang/rust public library APIs then please comment with @rustbot label +T-libs-api -T-libs to tag it appropriately. If this PR contains changes to any unstable APIs please edit the PR description to add a link to the relevant API Change Proposal or create one if you haven't already. If you're unsure where your change falls no worries, just leave it as is and the reviewer will take a look and make a decision to forward on if necessary.

Examples of T-libs-api changes:

Stabilizing library features
Introducing insta-stable changes such as new implementations of existing stable traits on existing stable types
Introducing new or changing existing unstable library APIs (excluding permanently unstable features / features without a tracking issue)
Changing public documentation in ways that create new stability guarantees
Changing observable runtime behavior of library APIs

workingjubilee · 2022-11-12T23:18:51Z

what busts where Pointee::Metadata = ()?

compiler-errors · 2022-11-12T23:23:00Z

@workingjubilee -- are you asking why this is enforcing something that's stricter than just T: Pointee<Metadata = ()>? Well, the currently implementation of dyn* treats dyn* Trait as being usize-sized, so we need to enforce that the thing that's being coerced into a dyn* is that size as well.

We could probably relax this to check size <= sizeof(usize), but I wanted to keep it simple. Also I need to do an alignment check I guess.

If that isn't what you're asking, then could you maybe re-explain?

compiler-errors · 2022-11-12T23:27:34Z

Unless you were asking what happens in the case like this:

#![feature(ptr_metadata)]
#![feature(dyn_star)]

use std::fmt::Debug;
use std::ptr::Thin;

fn polymorphic<T: Debug + ?Sized + Thin>(t: &T) {
    let _ = t as dyn* Debug;
}

In which case:

error[E0277]: `&T` needs to be a pointer-sized type
 --> /home/gh-compiler-errors/test.rs:9:13
  |
9 |     let _ = t as dyn* Debug;
  |             ^ `&T` needs to be a pointer-sized type
  |
  = help: the trait `PointerSized` is not implemented for `&T`
help: consider introducing a `where` clause, but there might be an alternative better way to express this requirement
  |
8 | fn polymorphic<T: Debug + ?Sized + Thin>(t: &T) where &T: PointerSized {
  |                                                 ++++++++++++++++++++++

This happens. Layout computation is not smart enough to take in account that T: Thin implies sizeof(&T) = sizeof(usize) 🤔

workingjubilee · 2022-11-13T03:14:37Z

Hmm, I'm not sure which I was asking, honestly, but I am satisfied with that answer, aside from the question of whether or not layout computation could simply be taught that instead I suppose. But that can be tabled for now.

compiler-errors · 2022-11-13T03:16:23Z

Yeah, I have a follow-up commit to teach layout computation exactly that, but I want to separate that from this PR since it maybe needs a bit more inspection.

bors · 2022-11-15T10:37:42Z

☔ The latest upstream changes (presumably #104428) made this pull request unmergeable. Please resolve the merge conflicts.

compiler/rustc_trait_selection/src/traits/select/candidate_assembly.rs

library/core/src/marker.rs

eholk · 2022-11-15T22:35:48Z

@bors r+

bors · 2022-11-15T22:35:50Z

📌 Commit 5edbf2c has been approved by eholk

It is now in the queue for this repository.

…holk Enforce that `dyn*` coercions are actually pointer-sized Implement a perma-unstable, rudimentary `PointerSized` trait to enforce `dyn*` casts are `usize`-sized for now, at least to prevent ICEs and weird codegen issues from cropping up after monomorphization since currently we enforce *nothing*. This probably can/should be removed in favor of a more sophisticated trait for handling `dyn*` conversions when we decide on one, but I just want to get something up for discussion and experimentation for now. r? `@eholk` cc `@tmandry` (though feel free to claim/reassign) Fixes rust-lang#102141 Fixes rust-lang#102173

compiler-errors · 2022-11-16T04:16:18Z

@bors r-

compiler-errors · 2022-11-16T06:13:28Z

@rustbot ready

eholk · 2022-11-17T23:26:13Z

@bors r+

bors · 2022-11-17T23:26:16Z

📌 Commit 4e20fa5209c868c48a6f044e76447bef004a87a7 has been approved by eholk

It is now in the queue for this repository.

bors · 2022-11-18T14:09:40Z

☔ The latest upstream changes (presumably #104555) made this pull request unmergeable. Please resolve the merge conflicts.

compiler-errors · 2022-11-18T18:59:41Z

Rebased past oli's Obligation constructor PR.

@bors r=eholk

bors · 2022-11-18T18:59:43Z

📌 Commit 39e076a has been approved by eholk

It is now in the queue for this repository.

…holk Enforce that `dyn*` coercions are actually pointer-sized Implement a perma-unstable, rudimentary `PointerSized` trait to enforce `dyn*` casts are `usize`-sized for now, at least to prevent ICEs and weird codegen issues from cropping up after monomorphization since currently we enforce *nothing*. This probably can/should be removed in favor of a more sophisticated trait for handling `dyn*` conversions when we decide on one, but I just want to get something up for discussion and experimentation for now. r? `@eholk` cc `@tmandry` (though feel free to claim/reassign) Fixes rust-lang#102141 Fixes rust-lang#102173

…holk Enforce that `dyn*` coercions are actually pointer-sized Implement a perma-unstable, rudimentary `PointerSized` trait to enforce `dyn*` casts are `usize`-sized for now, at least to prevent ICEs and weird codegen issues from cropping up after monomorphization since currently we enforce *nothing*. This probably can/should be removed in favor of a more sophisticated trait for handling `dyn*` conversions when we decide on one, but I just want to get something up for discussion and experimentation for now. r? ``@eholk`` cc ``@tmandry`` (though feel free to claim/reassign) Fixes rust-lang#102141 Fixes rust-lang#102173

…earth Rollup of 8 pull requests Successful merges: - rust-lang#102977 (remove HRTB from `[T]::is_sorted_by{,_key}`) - rust-lang#103378 (Fix mod_inv termination for the last iteration) - rust-lang#103456 (`unchecked_{shl|shr}` should use `u32` as the RHS) - rust-lang#103701 (Simplify some pointer method implementations) - rust-lang#104047 (Diagnostics `icu4x` based list formatting.) - rust-lang#104338 (Enforce that `dyn*` coercions are actually pointer-sized) - rust-lang#104498 (Edit docs for `rustc_errors::Handler::stash_diagnostic`) - rust-lang#104556 (rustdoc: use `code-header` class to format enum variants) Failed merges: r? `@ghost` `@rustbot` modify labels: rollup

…s-thin-ptr, r=wesleywiser layout_of: `T: Thin` implies `sizeof(&T) == sizeof(usize)` Use the `<T as Pointee>::Metadata` associated type to calculate the layout of a pointee's metadata, instead of hard-coding rules about certain types. Maybe this approach is overkill -- we could instead hard-code this approach as a fallback, with the matching on `Slice`/`Dynamic`/etc. happening first Fixes this issue here rust-lang#104338 (comment) .. But is also useful with transmutes, for example, given the UI test I added below.

rustbot assigned eholk Nov 12, 2022

rustbot added S-waiting-on-review Status: Awaiting review from the assignee but also interested parties. T-compiler Relevant to the compiler team, which will review and decide on the PR/issue. T-libs Relevant to the library team, which will review and decide on the PR/issue. labels Nov 12, 2022

compiler-errors mentioned this pull request Nov 13, 2022

layout_of: T: Thin implies sizeof(&T) == sizeof(usize) #104376

Merged

compiler-errors force-pushed the pointer-sized branch from 59c99fb to 84b06d5 Compare November 15, 2022 16:43

eholk reviewed Nov 15, 2022

View reviewed changes

compiler/rustc_trait_selection/src/traits/select/candidate_assembly.rs Outdated Show resolved Hide resolved

eholk reviewed Nov 15, 2022

View reviewed changes

library/core/src/marker.rs Show resolved Hide resolved

bors added S-waiting-on-bors Status: Waiting on bors to run and complete tests. Bors will change the label on completion. and removed S-waiting-on-review Status: Awaiting review from the assignee but also interested parties. labels Nov 15, 2022

matthiaskrgr mentioned this pull request Nov 16, 2022

Rollup of 12 pull requests #104474

Closed

This comment has been minimized.

Sign in to view

bors 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-bors Status: Waiting on bors to run and complete tests. Bors will change the label on completion. labels Nov 16, 2022

crlf0710 added the F-dyn_star `#![feature(dyn_star)]` label Nov 16, 2022

rustbot added S-waiting-on-review Status: Awaiting review from the assignee but also interested parties. and removed S-waiting-on-author Status: This is awaiting some action (such as code changes or more information) from the author. labels Nov 16, 2022

bors added S-waiting-on-bors Status: Waiting on bors to run and complete tests. Bors will change the label on completion. and removed S-waiting-on-review Status: Awaiting review from the assignee but also interested parties. labels Nov 17, 2022

bors 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-bors Status: Waiting on bors to run and complete tests. Bors will change the label on completion. labels Nov 18, 2022

compiler-errors added 3 commits November 18, 2022 18:23

Enforce that dyn* casts are actually pointer-sized

da3c539

Check both align and size in PointerSized

b3da04a

Only enforce ABI-mandated align, not preferred align is compatible

39e076a

compiler-errors force-pushed the pointer-sized branch from 4e20fa5 to 39e076a Compare November 18, 2022 18:24

bors added S-waiting-on-bors Status: Waiting on bors to run and complete tests. Bors will change the label on completion. and removed S-waiting-on-author Status: This is awaiting some action (such as code changes or more information) from the author. labels Nov 18, 2022

Manishearth mentioned this pull request Nov 18, 2022

Rollup of 9 pull requests #104585

Closed

This was referenced Nov 18, 2022

Rollup of 8 pull requests #104589

Closed

Rollup of 8 pull requests #104591

Merged

bors merged commit 24ee599 into rust-lang:master Nov 19, 2022

rustbot added this to the 1.67.0 milestone Nov 19, 2022

compiler-errors deleted the pointer-sized branch August 11, 2023 20:13

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

Enforce that `dyn*` coercions are actually pointer-sized #104338

Enforce that `dyn*` coercions are actually pointer-sized #104338

compiler-errors commented Nov 12, 2022 •

edited

Loading

rustbot commented Nov 12, 2022

workingjubilee commented Nov 12, 2022

compiler-errors commented Nov 12, 2022 •

edited

Loading

compiler-errors commented Nov 12, 2022 •

edited

Loading

workingjubilee commented Nov 13, 2022

compiler-errors commented Nov 13, 2022

bors commented Nov 15, 2022

eholk commented Nov 15, 2022

bors commented Nov 15, 2022

This comment has been minimized.

compiler-errors commented Nov 16, 2022

compiler-errors commented Nov 16, 2022

eholk commented Nov 17, 2022

bors commented Nov 17, 2022

bors commented Nov 18, 2022

compiler-errors commented Nov 18, 2022

bors commented Nov 18, 2022

Enforce that dyn* coercions are actually pointer-sized #104338

Enforce that dyn* coercions are actually pointer-sized #104338

Conversation

compiler-errors commented Nov 12, 2022 • edited Loading

rustbot commented Nov 12, 2022

workingjubilee commented Nov 12, 2022

compiler-errors commented Nov 12, 2022 • edited Loading

compiler-errors commented Nov 12, 2022 • edited Loading

workingjubilee commented Nov 13, 2022

compiler-errors commented Nov 13, 2022

bors commented Nov 15, 2022

eholk commented Nov 15, 2022

bors commented Nov 15, 2022

This comment has been minimized.

compiler-errors commented Nov 16, 2022

compiler-errors commented Nov 16, 2022

eholk commented Nov 17, 2022

bors commented Nov 17, 2022

bors commented Nov 18, 2022

compiler-errors commented Nov 18, 2022

bors commented Nov 18, 2022

Enforce that `dyn*` coercions are actually pointer-sized #104338

Enforce that `dyn*` coercions are actually pointer-sized #104338

compiler-errors commented Nov 12, 2022 •

edited

Loading

compiler-errors commented Nov 12, 2022 •

edited

Loading

compiler-errors commented Nov 12, 2022 •

edited

Loading