compiler-builtins subtree update by tgross35 · Pull Request #162816 · rust-lang/rust

tgross35 · 2026-09-15T15:28:55Z

Subtree update of compiler-builtins to rust-lang/compiler-builtins@9e64861.

Created using https://github.com/rust-lang/josh-sync.

This target was downgraded to Tier 2 without host tools [1]. Closes: rust-lang/compiler-builtins#1306 [1]: rust-lang/compiler-team#1020

Note that target features on ARM are not yet stable (tracked by [1]) so this only actually gets used on nightly. [1]: rust-lang#150246 [ add commit body - Trevor ]

In LLVM 23, calls to `wcslen` are sometimes generated by a loop optimization pass: llvm/llvm-project#132572. This causes some Rust targets (including but not limited to the UEFI targets) to fail with a linker error if a loop is transformed into a `wcslen` call. Provide a simple implementation of `wcslen` in compiler-builtins to fix this. Fixes rust-lang#160827

`f16` and `f128` have abi issues, so continue to use `extern "Rust"` for now

Introduce a constant for the result of perfoming a quietening operation on our default sNaN. Additionally check in the C file used to retrieve these values.

This is slightly more robust against potential issues from overriding the builtin symbol.

Provide reasoning for skips, and enable thumb since this seems to work. Also, change wasm32-unknown-unknown to be a proper build-only target. The mentioned failures on MinGW are cases like: thread 'musl_edge_case_expf' (1728) panicked at libm-test\tests\compare_built_musl.rs:26:50: called `Result::unwrap()` on an `Err` value: input: (0.0,) as hex: (+0x0p+0,) as bits: (0x00000000,) expected: 0.0 +0x0p+0 0x00000000 actual: 1.0 +0x1.000000p+0 0x3f800000 thread 'musl_edge_case_ceil' (1964) panicked at libm-test\tests\compare_built_musl.rs:26:50: called `Result::unwrap()` on an `Err` value: input: (1.0000000000000002,) as hex: (+0x1.0000000000001p+0,) as bits: (0x3ff0000000000001,) expected: 1.0 +0x1.0000000000000p+0 0x3ff0000000000000 actual: 2.0 +0x1.0000000000000p+1 0x4000000000000000 Our libm's results are correct, for some reason musl is off (something about the fp environment on x86, or a different FLT_EVAL_MODE?)

This updates the rust-version file to 32d94cc.

Pull recent changes from https://github.com/rust-lang/rust via Josh. Upstream ref: rust-lang/rust@32d94cc Filtered ref: rust-lang/compiler-builtins@549dea8 Upstream diff: rust-lang/rust@45f215f...32d94cc This merge was created using https://github.com/rust-lang/josh-sync.

This is a rather old config, added in c924aed0b9ac ("Fix Armv8-M Baseline compilation"). `thumb_1` is somewhat misleading, since targets with `thumb2` also support basic `thumb` instructions. Rename it to be more clear about what this is actually checking. The config is not currently used but may be in the future.

We still match on target name because we can't always rely on config from the unstable `arm_target_feature`. However, we can still assert that things match up when running in the compiler-builtins CI. This is disabled for now because it fails on three thumb targets: * thumbv4t-none-eabi * thumbv5te-none-eabi * thumbv6-none-eabi This will be resolved in a future commit. Checking for the verbose build is a good indicator that we are in our CI, rather than any other that might set the `CI` env.

Fixes: rust-lang#162259

Previously CI would only run tests for ARM mode, Thumb mode targets were compile-only. This ensures Thumb2 is covered. Thumb1 is still not covered, but that's less easy since there's no Linux Thumb1 target. This commit fixes some places that assume "thumb" means "bare metal". They now check for `target_os = "none"` instead.

tgross35 · 2026-09-15T15:37:09Z

@bors try jobs=test-various,test-armhf-gnu,dist-arm-linux-gnueabi,dist-armhf-linux,dist-armv7-linux

compiler-builtins subtree update try-job: test-various try-job: test-armhf-gnu try-job: dist-arm-linux-gnueabi try-job: dist-armhf-linux try-job: dist-armv7-linux

tgross35 · 2026-09-15T17:34:45Z

@bors r+ rollup=never p=1

rust-bors · 2026-09-15T17:34:48Z

📌 Commit 8864022 has been approved by tgross35

It is now in the queue for this repository.

rust-bors · 2026-09-15T17:46:23Z

☀️ Try build successful (CI)
Build commit: 79e7068 (79e7068281d285556f3145941ae62245c7ad8d3f)
Base parent: 5392d2f (5392d2f545c6836dc79f209bcc14ac7179dbc4f2)

tgross35 · 2026-09-15T19:12:38Z

@rustbot label +beta-nominated

Looking to backport a7f33f3 and 192ea4f to fix #162259. Since the LLVM update (on beta) that issue shows up on arm targets that we don't build C versions of intrinsics for, which at least includes some no-std thumb targets.

Two other trivial commits are needed to avoid a conflict, I've prepared this in #162820.

rust-bors · 2026-09-15T22:11:46Z

☀️ Test successful - CI
Approved by: tgross35
Duration: 3h 13m 7s
Pushing 215a8af to main...

github-actions · 2026-09-15T22:15:09Z

What is this?

This is an experimental post-merge analysis report that shows differences in test outcomes between the merged PR and its parent PR.

Comparing 0da8aed (parent) -> 215a8af (this PR)

Test differences

Show 4 test diffs

4 doctest diffs were found. These are ignored, as they are noisy.

Test dashboard

Run

cargo run --manifest-path src/ci/citool/Cargo.toml -- \
    test-dashboard 215a8af4bb4c106cccf6d6535f84eaae91818265 --output-dir test-dashboard

And then open test-dashboard/index.html in your browser to see an overview of all executed tests.

Job duration changes

dist-x86_64-llvm-mingw: 1h 19m -> 2h 11m (+65.3%)
test-x86_64-gnu-gcc-core-tests: 17m 39s -> 9m 19s (-47.2%)
dist-ohos-x86_64: 57m 17s -> 1h 23m (+46.3%)
test-x86_64-gnu-distcheck: 1h 38m -> 2h 20m (+42.7%)
test-x86_64-gnu: 1h 52m -> 2h 39m (+42.2%)
dist-loongarch64-linux: 1h 28m -> 2h 4m (+41.0%)
dist-i686-linux: 1h 21m -> 1h 53m (+39.2%)
test-x86_64-gnu-next-trait-solver-polonius: 42m 15s -> 56m 29s (+33.7%)
dist-arm-linux-musl: 1h 26m -> 1h 54m (+32.6%)
test-x86_64-gnu-stdlib-semver-check: 16m 39s -> 11m 18s (-32.1%)

How to interpret the job duration changes?

Job durations can vary a lot, based on the actual runner instance
that executed the job, system noise, invalidated caches, etc. The table above is provided
mostly for t-infra members, for simpler debugging of potential CI slow-downs.

rust-timer · 2026-09-15T22:57:54Z

Finished benchmarking commit (215a8af): comparison URL.

Overall result: ❌ regressions - no action needed

@rustbot label: -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)	0.3%	[0.2%, 0.4%]	2
Improvements ✅ (primary)	-	-	0
Improvements ✅ (secondary)	-	-	0
All ❌✅ (primary)	-	-	0

Max RSS (memory usage)

Results (primary 0.7%, secondary -2.5%)

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

	mean	range	count
Regressions ❌ (primary)	0.7%	[0.7%, 0.7%]	1
Regressions ❌ (secondary)	8.2%	[8.2%, 8.2%]	1
Improvements ✅ (primary)	-	-	0
Improvements ✅ (secondary)	-4.3%	[-7.0%, -2.3%]	6
All ❌✅ (primary)	0.7%	[0.7%, 0.7%]	1

Cycles

Results (primary 4.2%, secondary 0.3%)

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

	mean	range	count
Regressions ❌ (primary)	4.2%	[2.8%, 5.6%]	2
Regressions ❌ (secondary)	5.0%	[3.0%, 8.1%]	5
Improvements ✅ (primary)	-	-	0
Improvements ✅ (secondary)	-5.7%	[-7.8%, -2.0%]	4
All ❌✅ (primary)	4.2%	[2.8%, 5.6%]	2

Binary size

Results (primary 0.0%, secondary 0.0%)

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

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

Bootstrap: 494.609s -> 498.703s (0.83%)
Artifact size: 406.84 MiB -> 406.94 MiB (0.03%)

rustbot · 2026-09-17T10:13:39Z

beta backport approved as per libs team on Zulip. A backport PR will be authored by the release team at the end of the current development cycle. Backport labels are handled by them.

[beta] backports - Re-export `core::fmt::NumBuffer` in `alloc` (and `std`) #161430 - Make `VaArgSafe` dyn-incompatible #162374 - Destabilize `VaArgSafe` #162909 - attach global target features to module-level assembly #160594 - (partial) compiler-builtins subtree update - #162816 r? me

tgross35 and others added 18 commits September 1, 2026 14:38

ci: Rename channel to host_toolchain

a657aee

ci: Switch i686-pc-windows-gnu to a 64-bit host and re-enable

6414f1c

This target was downgraded to Tier 2 without host tools [1]. Closes: rust-lang/compiler-builtins#1306 [1]: rust-lang/compiler-team#1020

libm: support hardware sqrtf/sqrt on Arm32 and Thumb targets

ce380fa

Note that target features on ARM are not yet stable (tracked by [1]) so this only actually gets used on nightly. [1]: rust-lang#150246 [ add commit body - Trevor ]

use extern "C" and C names for complex div and mul of f32 and f64

a14f914

`f16` and `f128` have abi issues, so continue to use `extern "Rust"` for now

libm: Add a quieted signaling NaN constant

67824fb

Introduce a constant for the result of perfoming a quietening operation on our default sNaN. Additionally check in the C file used to retrieve these values.

libm: Make assert_biteq usable outside of libm

2df9604

test: Replace wcslen no_mangle wrapper with optimize(none)

006120f

This is slightly more robust against potential issues from overriding the builtin symbol.

Prepare for merging from rust-lang/rust

6d30d7c

This updates the rust-version file to 32d94cc.

c-b: Preserve NaN signs in soft-float addition

8242188

test: Deduplicate addition and subtraction NaN tests, improve output

60ab542

c-b: Add aeabi_u(read|write)(4|8) intrinsics.

a7f33f3

Fixes: rust-lang#162259

build: Fix thumb_1 cfg to include all "thumb but not thumbv2" targets.

192ea4f

This comment has been minimized.

Sign in to view

rust-bors Bot 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 Sep 15, 2026

This comment has been minimized.

Sign in to view

tgross35 mentioned this pull request Sep 15, 2026

[beta] Backports for compiler-builtins #162820

Closed

rustbot added the beta-nominated Nominated for backporting to the compiler in the beta channel. label Sep 15, 2026

rust-bors Bot added merged-by-bors This PR was explicitly merged by bors. and removed S-waiting-on-bors Status: Waiting on bors to run and complete tests. Bors will change the label on completion. labels Sep 15, 2026

rust-bors Bot merged commit 215a8af into rust-lang:main Sep 15, 2026
15 checks passed

rustbot added this to the 1.100.0 milestone Sep 15, 2026

tgross35 deleted the update-builtins branch September 15, 2026 22:38

This was referenced Sep 16, 2026

rust-lld: error: undefined symbol: wcslen rust-osdev/bootloader#579

Closed

Provide wcslen for x86_64-unknown-uefi target rust-osdev/bootloader#582

Closed

rustbot added the beta-accepted Accepted for backporting to the compiler in the beta channel. label Sep 17, 2026

cuviper mentioned this pull request Sep 19, 2026

[beta] backports #162987

Merged

cuviper removed the beta-nominated Nominated for backporting to the compiler in the beta channel. label Sep 19, 2026

crawfxrd mentioned this pull request Sep 21, 2026

Build Standard Binary job is currently failing rust-osdev/uefi-rs#2018

Closed

cuviper mentioned this pull request Oct 1, 2026

Regression in Rust 1.99: missing symbol wcslen #163614

Open

github-actions Bot mentioned this pull request Oct 2, 2026

Toolchain upgrade to nightly-2026-09-25 flux-rs/flux#1803

Merged

Navigation Menu

Uh oh!

compiler-builtins subtree update - #162816

compiler-builtins subtree update#162816
rust-bors[bot] merged 18 commits into
rust-lang:mainfrom
tgross35:update-builtins

tgross35 commented Sep 15, 2026

Uh oh!

tgross35 commented Sep 15, 2026

Uh oh!

This comment has been minimized.

tgross35 commented Sep 15, 2026

Uh oh!

rust-bors Bot commented Sep 15, 2026

Uh oh!

rust-bors Bot commented Sep 15, 2026

Uh oh!

This comment has been minimized.

tgross35 commented Sep 15, 2026

Uh oh!

rust-bors Bot commented Sep 15, 2026

Uh oh!

Uh oh!

github-actions Bot commented Sep 15, 2026

Uh oh!

rust-timer commented Sep 15, 2026

Uh oh!

rustbot commented Sep 17, 2026

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

9 participants

Uh oh!

Conversation

tgross35 commented Sep 15, 2026

Uh oh!

tgross35 commented Sep 15, 2026

Uh oh!

This comment has been minimized.

tgross35 commented Sep 15, 2026

Uh oh!

rust-bors Bot commented Sep 15, 2026

Uh oh!

rust-bors Bot commented Sep 15, 2026

Uh oh!

This comment has been minimized.

tgross35 commented Sep 15, 2026

Uh oh!

rust-bors Bot commented Sep 15, 2026

Uh oh!

Uh oh!

github-actions Bot commented Sep 15, 2026

Test differences

Job duration changes

Uh oh!

rust-timer commented Sep 15, 2026

Overall result: ❌ regressions - no action needed

Instruction count

Max RSS (memory usage)

Cycles

Binary size

Uh oh!

rustbot commented Sep 17, 2026

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

9 participants