make volatile operations const by RalfJung · Pull Request #159092 · rust-lang/rust

RalfJung · 2026-07-10T14:52:34Z

This came up in gfx-rs/wgpu#8897 (comment). The tl;dr is that it can be useful to have shared codepaths between things that need to work on special memory (and hence need to use volatile) and things that work on regular Rust memory, and for the regular Rust memory case those things totally could and should be const fn.

So, let's (unstably, for now) allow volatile reads and writes in const, and specify them to behave just like normal reads and writes. In const we know that there's nobody trying to observe those locations so there's nothing special we have to do with those accesses.

Cc @rust-lang/opsem @rust-lang/wg-const-eval

Tracking issue: #159094

rustbot · 2026-07-10T14:52:37Z

miri is developed in its own repository. If possible, consider making this change to rust-lang/miri instead.

cc @rust-lang/miri

Some changes occurred to the intrinsics. Make sure the CTFE / Miri interpreter
gets adapted for the changes, if necessary.

cc @rust-lang/miri, @oli-obk, @lcnr

Some changes occurred to the CTFE machinery

cc @oli-obk, @lcnr

Some changes occurred to the CTFE / Miri interpreter

cc @rust-lang/miri

rustbot · 2026-07-10T14:52:40Z

r? @JohnTitor

rustbot has assigned @JohnTitor.
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: libs
libs expanded to 12 candidates
Random selection from 6 candidates

oli-obk · 2026-07-10T18:07:34Z

                self.typed_swap_nonoverlapping_intrinsic(&args[0], &args[1])?;
            }

+            sym::volatile_load => {


This is the shared space for intrinsics for miri and ctfe. But this impl will only be used by ctfe, so it should live in the ctfe intrinsic impl file

View changes since the review

No, the same impl should be used by CTFE and Miri.

Ah I assumed we'd want do so extra checks in miri, although who but other threads would be randomly modifying memory idk either

oli-obk · 2026-07-10T20:09:35Z

r? @oli-obk

@bors r+

rust-bors · 2026-07-10T20:09:39Z

📌 Commit d9e15dd has been approved by oli-obk

It is now in the queue for this repository.

oli-obk · 2026-07-10T20:09:41Z

@bors rollup

make volatile operations const This came up in gfx-rs/wgpu#8897 (comment). The tl;dr is that it can be useful to have shared codepaths between things that need to work on special memory (and hence need to use volatile) and things that work on regular Rust memory, and for the regular Rust memory case those things totally could and should be `const fn`. So, let's (unstably, for now) allow volatile reads and writes in `const`, and specify them to behave just like normal reads and writes. In const we *know* that there's nobody trying to observe those locations so there's nothing special we have to do with those accesses. Cc @rust-lang/opsem @rust-lang/wg-const-eval Tracking issue: rust-lang#159094

Rollup of 15 pull requests Successful merges: - #159106 (Update LLVM) - #157690 (codegen_ssa: pack small const aggregates into immediate stores) - #159005 (Use `as_lang_item` instead of repeatedly matching) - #156735 (Move NativeLib::filename to the rmeta-link archive member) - #157153 (allow `Allocator`s to be used as `#[global_allocator]`s) - #158269 (jq directives for jsondocck) - #158767 (merge DefKind::InlineConst into AnonConst) - #159099 (Stabilize String::from_utf8_lossy_owned) - #158930 (Reorganize `tests/ui/issues` [20/N]) - #158965 (Add codegen test for Result is_ok unwrap) - #158979 (Reorganize `tests/ui/issues` [21/N]) - #159050 (assert only opaques with sub unified hidden infer are non-rigid) - #159062 (Remove unused WEAK_ONLY_LANG_ITEMS static) - #159070 (Add `io::ErrorKind::InputOutputError`) - #159092 (make volatile operations const)

Rollup merge of #159092 - RalfJung:const-volatile, r=oli-obk make volatile operations const This came up in gfx-rs/wgpu#8897 (comment). The tl;dr is that it can be useful to have shared codepaths between things that need to work on special memory (and hence need to use volatile) and things that work on regular Rust memory, and for the regular Rust memory case those things totally could and should be `const fn`. So, let's (unstably, for now) allow volatile reads and writes in `const`, and specify them to behave just like normal reads and writes. In const we *know* that there's nobody trying to observe those locations so there's nothing special we have to do with those accesses. Cc @rust-lang/opsem @rust-lang/wg-const-eval Tracking issue: #159094

JonathanBrouwer · 2026-07-12T07:38:53Z

@rust-timer build 95ccb43

Rollup of 15 pull requests Successful merges: - rust-lang/rust#159106 (Update LLVM) - rust-lang/rust#157690 (codegen_ssa: pack small const aggregates into immediate stores) - rust-lang/rust#159005 (Use `as_lang_item` instead of repeatedly matching) - rust-lang/rust#156735 (Move NativeLib::filename to the rmeta-link archive member) - rust-lang/rust#157153 (allow `Allocator`s to be used as `#[global_allocator]`s) - rust-lang/rust#158269 (jq directives for jsondocck) - rust-lang/rust#158767 (merge DefKind::InlineConst into AnonConst) - rust-lang/rust#159099 (Stabilize String::from_utf8_lossy_owned) - rust-lang/rust#158930 (Reorganize `tests/ui/issues` [20/N]) - rust-lang/rust#158965 (Add codegen test for Result is_ok unwrap) - rust-lang/rust#158979 (Reorganize `tests/ui/issues` [21/N]) - rust-lang/rust#159050 (assert only opaques with sub unified hidden infer are non-rigid) - rust-lang/rust#159062 (Remove unused WEAK_ONLY_LANG_ITEMS static) - rust-lang/rust#159070 (Add `io::ErrorKind::InputOutputError`) - rust-lang/rust#159092 (make volatile operations const)

rust-timer · 2026-07-12T16:23:03Z

Finished benchmarking commit (95ccb43): comparison URL.

Overall result: ❌✅ regressions and improvements - please read:

Benchmarking means the PR may be perf-sensitive. It's automatically marked not fit for rolling up. Overriding is possible but disadvised: it risks changing compiler perf.

Next, please: If you can, justify the regressions found in this try perf run in writing along with @rustbot label: +perf-regression-triaged. If not, fix the regressions and do another perf run. Neutral or positive results will clear the label automatically.

@bors rollup=never
@rustbot label: -S-waiting-on-perf +perf-regression

Instruction count

Our most reliable metric. Used to determine the overall result above. However, even this metric can be noisy.

	mean	range	count
Regressions ❌ (primary)	-	-	0
Regressions ❌ (secondary)	0.0%	[0.0%, 0.1%]	3
Improvements ✅ (primary)	-	-	0
Improvements ✅ (secondary)	-0.1%	[-0.2%, -0.0%]	7
All ❌✅ (primary)	-	-	0

Max RSS (memory usage)

Results (secondary 1.3%)

A less reliable metric. May be of interest, but not used to determine the overall result above.

	mean	range	count
Regressions ❌ (primary)	-	-	0
Regressions ❌ (secondary)	3.5%	[3.5%, 3.5%]	1
Improvements ✅ (primary)	-	-	0
Improvements ✅ (secondary)	-0.9%	[-0.9%, -0.9%]	1
All ❌✅ (primary)	-	-	0

Cycles

Results (primary -2.3%, secondary 1.7%)

A less reliable metric. May be of interest, but not used to determine the overall result above.

	mean	range	count
Regressions ❌ (primary)	-	-	0
Regressions ❌ (secondary)	5.4%	[2.5%, 9.3%]	4
Improvements ✅ (primary)	-2.3%	[-2.3%, -2.3%]	1
Improvements ✅ (secondary)	-3.2%	[-4.0%, -2.7%]	3
All ❌✅ (primary)	-2.3%	[-2.3%, -2.3%]	1

Binary size

This perf run didn't have relevant results for this metric.

Bootstrap: 490.729s -> 491.445s (0.15%)
Artifact size: 389.29 MiB -> 389.91 MiB (0.16%)

JonathanBrouwer · 2026-07-12T20:05:20Z

@rustbot label: +perf-regression-triaged
Noise

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 Jul 10, 2026

rustbot assigned JohnTitor Jul 10, 2026

RalfJung mentioned this pull request Jul 10, 2026

Tracking Issue for const_volatile #159094

Open

4 tasks

make volatile operations const

d9e15dd

RalfJung force-pushed the const-volatile branch from c542cd8 to d9e15dd Compare July 10, 2026 14:59

RalfJung mentioned this pull request Jul 10, 2026

BufferViewMut is unsound due to WC memory gfx-rs/wgpu#8897

Open

oli-obk reviewed Jul 10, 2026

View reviewed changes

rustbot assigned oli-obk Jul 10, 2026

rust-bors Bot added the S-waiting-on-bors Status: Waiting on bors to run and complete tests. Bors will change the label on completion. label Jul 10, 2026

rustbot unassigned JohnTitor Jul 10, 2026

rust-bors Bot removed the S-waiting-on-review Status: Awaiting review from the assignee but also interested parties. label Jul 10, 2026

JonathanBrouwer mentioned this pull request Jul 10, 2026

Rollup of 14 pull requests #159104

Closed

JonathanBrouwer mentioned this pull request Jul 10, 2026

Rollup of 15 pull requests #159107

Closed

jhpratt mentioned this pull request Jul 11, 2026

Rollup of 15 pull requests #159113

Merged

rust-bors Bot merged commit 3657ecd into rust-lang:main Jul 11, 2026
13 checks passed

rustbot added this to the 1.99.0 milestone Jul 11, 2026

RalfJung deleted the const-volatile branch July 11, 2026 08:33

This comment has been minimized.

Sign in to view

rustbot added the perf-regression Performance regression. label Jul 12, 2026

rustbot added the perf-regression-triaged The performance regression has been triaged. label Jul 12, 2026

Uh oh!

Uh oh!

Conversation

RalfJung commented Jul 10, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

rustbot commented Jul 10, 2026

Uh oh!

rustbot commented Jul 10, 2026

Uh oh!

oli-obk Jul 10, 2026 • 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!

RalfJung Jul 10, 2026

Choose a reason for hiding this comment

Uh oh!

oli-obk Jul 10, 2026

Choose a reason for hiding this comment

Uh oh!

oli-obk commented Jul 10, 2026

Uh oh!

rust-bors Bot commented Jul 10, 2026

Uh oh!

oli-obk commented Jul 10, 2026

Uh oh!

Uh oh!

JonathanBrouwer commented Jul 12, 2026

Uh oh!

This comment has been minimized.

rust-timer commented Jul 12, 2026

Overall result: ❌✅ regressions and improvements - please read:

Instruction count

Max RSS (memory usage)

Cycles

Binary size

Uh oh!

JonathanBrouwer commented Jul 12, 2026

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

6 participants

RalfJung commented Jul 10, 2026 •

edited

Loading

oli-obk Jul 10, 2026 •

edited by rustbot

Loading