compiler-builtins subtree update by tgross35 · Pull Request #153866 · rust-lang/rust

tgross35 · 2026-03-14T12:36:41Z

Subtree update of compiler-builtins to rust-lang/compiler-builtins@8173070.

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

It would be nice to support a few more flags here, but the current implementation is a bit limited. Switch to `getopts` which should make things a bit easier in the future.

PE executables don't seem to have anything in the symbol table. Don't assert that we find any symbols in this case, which allows using symcheck for executable binaries.

Implement the following logic: * For elf executable binaries, check for PT_GNU_STACK with PF_X. This combination tells the kernel to make the stack executable. * For elf intermediate objects, check for `.note.GNU-stack` with SHF_EXECINSTR. This combination in an object file tells the linker to give the final binary PT_GNU_STACK PF_X. * For elf intermediate objects with no `.note.GNU-stack`, assume the legacy behavior that assumes an executable stack is required. * For non-elf binaries, don't check anything. In a follow up it may be possible to check for `MH_ALLOW_STACK_EXECUTION` on Mach-O binaries, but it doesn't seem possible to get the latest compiler to emit this. This appears to match what is done by `scanelf` to emit `!WX` [1], which seems to be the tool used to create the output in the issue. The ld manpage [2] also has some useful notes about these flags, as does the presentation at [3]. Closes: rust-lang/compiler-builtins#183 [1]: https://github.com/gentoo/pax-utils/blob/9ef54b472e42ba2c5479fbd86b8be2275724b064/scanelf.c [2]: https://man7.org/linux/man-pages/man1/ld.1.html [3]: https://www.ndss-symposium.org/wp-content/uploads/6D-s0924-ye.pdf

CI is failing with: error: failed to run custom build command for `quote v1.0.44` Caused by: process didn't exit successfully: `/home/runner/.rustup/toolchains/nightly-x86_64-unknown-linux-gnu/bin/cargo-miri runner /home/runner/work/compiler-builtins/compiler-builtins/target/miri/debug/build/quote-32cc00414fa6f72d/build-script-build` (exit status: 1) --- stderr fatal error: file "/home/runner/work/compiler-builtins/compiler-builtins/target/miri/debug/build/quote-32cc00414fa6f72d/build-script-build" contains outdated or invalid JSON; try `cargo clean` Link: https://rust-lang.zulipchat.com/#narrow/channel/269128-miri/topic/build-script-build.20contains.20outdated.20or.20invalid.20JSON/with/573426109

These were disabled due to permission issues, which should now be resolved. This reverts commit 28e0556a7bca6ca828186d9de74d30c7cb45db5e.

The nightly has the LLVM bump that resolves the relevant issue. This reverts commit 99d1fc7752b0e09cff5729cd4f3783d23c23cb66. Link: rust-lang#152428

The `frexpf16` signature is added (to make traits easier) but left unimplemented, since it will need a slightly different algorithm.

This will support `frexpf16` and should allow `generic::frexp` to be inlined into the non-generic `frexp*` functions. Additionally remove an unneeded `as` cast.

The new algorithm now correctly supports the type.

This includes the most recent LLVM bump.

Values like 0.5 that have an exponent but not a mantissa are currently being reported as OK rather than INEXACT for ceil, floor, and trunc. Fix this and clean up the test cases. This tries to keep the algorithm diff as simple as possible, which introduces some small performance regressions. This is improved in a future commit.

ceil seems to optimize better, but this gets closer to the pre-fix codegen.

Suggested-by: Juho Kahala <57393910+quaternic@users.noreply.github.com>

Most of the semver-breaking changes here relate to the 0.10 release of `rand`. Changelogs: * https://github.com/criterion-rs/criterion.rs/releases/tag/criterion-v0.8.0 * https://github.com/gimli-rs/object/blob/master/CHANGELOG.md#0380 * https://github.com/rust-random/rand/blob/master/CHANGELOG.md#0100---2026-02-08

Currently Criterion is always built with the test crates, but it doesn't really need to be. It wasn't a problem before but did start showing up with the latest version bump since it added a C dependency, and for the Miri CI runs we don't have the C toolchains. To resolve this and speed up compilation time, move criterion behind a new feature `walltime`. I'd rather have named this `runtime` but don't want to confuse my future self with "what runtime?".

The tests pass successfully for me on powerpc64le-unknown-linux-gnu with nightly 1.95.

This was missed in 6e91b0346c3d (“Revert "Disable broken powerpc64 test due to https://github.com/rust-lang/rust/issues/88520"”).

For cases where we would like to run tests that aren't automatically detected, allow the following syntax in PR descriptions: ci: extra-extensive=copysignf,sqrtf16

In order to improve upon some inconsistencies seen across CI runs and locally, add a Cargo.lock file for the main workspace.

PRs are now getting the following: Traceback (most recent call last): File "/home/runner/work/compiler-builtins/compiler-builtins/ci/ci-util.py", line 510, in <module> main() File "/home/runner/work/compiler-builtins/compiler-builtins/ci/ci-util.py", line 496, in main ctx.emit_workflow_output() File "/home/runner/work/compiler-builtins/compiler-builtins/ci/ci-util.py", line 294, in emit_workflow_output pr = PrInfo.from_env() ^^^^^^^^^^^^^^^^^ File "/home/runner/work/compiler-builtins/compiler-builtins/ci/ci-util.py", line 152, in from_env return cls.from_pr(pr_env) ^^^^^^^^^^^^^^^^^^^ File "/home/runner/work/compiler-builtins/compiler-builtins/ci/ci-util.py", line 174, in from_pr return cls(**json.loads(pr_info), cfg=PrCfg(pr_json["body"])) ^^^^^^^^^^^^^^^^^^^^^^ File "/home/runner/work/compiler-builtins/compiler-builtins/ci/ci-util.py", line 134, in __init__ pprint.pp(self) File "/usr/lib/python3.12/pprint.py", line 66, in pp pprint(object, *args, sort_dicts=sort_dicts, **kwargs) ... AttributeError: 'PrCfg' object has no attribute 'extra_extensive'. Did you mean: 'skip_extensive'? Resolve this by using `__post_init__` rather than `__init__`. Fixes: bba024d20464 ("ci: Allow specifying extra extensive tests to run")

As of indicatif 0.18.4, this option must be enabled in able to build.

These warnings are showing up on the most recent nightly: error: feature `f128` is declared but not used --> builtins-test/tests/addsub.rs:3:35 | 3 | #![cfg_attr(f128_enabled, feature(f128))] | ^^^^ It isn't always easy to be more exact with the features we enable, so mostly ignore this.

Since we are starting to use the `fmaximum_num` intrinsics, LLVM may emit these.

Add a wrapper similar to `Hexf` to print integers with proper padding.

Make `Float::NAN` a qNaN, add sNaNs, and add operations to check whether a NaN is quiet or signaling. The MIPS values are checked against https://gcc.godbolt.org/z/1bhsd5ana.

Give more descriptive output in tests since we sometimes need to treat these differently. We still don't parse `sNaN`/`qNaN` for now, though we could in the future.

We do handle signaling NaNs properly, with the exception of raising exceptions as IEEE 754 requires. Add tests to this effect for `fmin`, `fminimum`, `fminimum_num`, and the max variants.

rustbot · 2026-03-14T12:36:48Z

⚠️ Warning ⚠️

This PR is based on an upstream commit that is older than 28 days.

It's recommended to update your branch according to the rustc-dev-guide.

RalfJung · 2026-03-14T12:57:39Z

Be very careful with force-pushing sync branches, this can break the round-trip. Do not modify or rebase any commits Josh created. Adding more commits on top is fine.

This PR is based on an upstream commit that is older than 28 days.

This indicates there hasn't been a sync the other direction in a while. In Miri we have those fully automated, happy to share that infrastructure if you are interested.

tgross35 · 2026-03-14T13:13:31Z

Be very careful with force-pushing sync branches, this can break the round-trip. Do not modify or rebase any commits Josh created. Adding more commits on top is fine.

I accidentally started the sync from the wrong branch and this was the correction, see rust-lang/josh-sync#36.

This PR is based on an upstream commit that is older than 28 days.

This indicates there hasn't been a sync the other direction in a while. In Miri we have those fully automated, happy to share that infrastructure if you are interested.

We have rust->cb syncs on a cron job and I also launched one before doing this, there just isn’t anything to sync. Unless you’re talking about doing this cb->rust sync automatically?

RalfJung · 2026-03-14T13:20:51Z

We have rust->cb syncs on a cron job and I also launched one before doing this, there just isn’t anything to sync.

Ah, yeah that would also explain the old base commit. :)

tgross35 · 2026-03-14T14:56:27Z

@bors r+ rollup=never p=1

rust-bors · 2026-03-14T14:56:31Z

📌 Commit 7a698be has been approved by tgross35

It is now in the queue for this repository.

rust-bors · 2026-03-14T20:07:50Z

☀️ Test successful - CI
Approved by: tgross35
Duration: 3h 11m 18s
Pushing 35f1109 to main...

github-actions · 2026-03-14T20:10:37Z

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 798a011 (parent) -> 35f1109 (this PR)

Test differences

Show 2 test diffs

2 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 35f1109ece895199e4b8d755b101e9a481d18243 --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-apple-various: 1h 23m -> 1h 40m (+19.8%)
i686-gnu-1: 2h 22m -> 1h 57m (-18.2%)
dist-armv7-linux: 1h 25m -> 1h 37m (+14.1%)
pr-check-1: 32m 51s -> 28m 51s (-12.2%)
i686-gnu-2: 1h 44m -> 1h 31m (-12.1%)
aarch64-msvc-1: 2h 19m -> 2h 3m (-11.5%)
i686-gnu-nopt-1: 2h 21m -> 2h 5m (-11.5%)
x86_64-gnu-stable: 2h 34m -> 2h 19m (-10.0%)
x86_64-gnu-debug: 2h 5m -> 1h 53m (-9.4%)
aarch64-gnu-llvm-21-2: 49m 56s -> 45m 29s (-8.9%)

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 · 2026-03-14T20:48:49Z

Finished benchmarking commit (35f1109): comparison URL.

Overall result: ❌✅ regressions and improvements - 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)	0.0%	[0.0%, 0.0%]	1
Improvements ✅ (primary)	-	-	0
Improvements ✅ (secondary)	-0.1%	[-0.1%, -0.1%]	1
All ❌✅ (primary)	-	-	0

Max RSS (memory usage)

Results (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
Regressions ❌ (secondary)	0.7%	[0.7%, 0.7%]	1
Improvements ✅ (primary)	-	-	0
Improvements ✅ (secondary)	-5.7%	[-5.7%, -5.7%]	1
All ❌✅ (primary)	-	-	0

Cycles

This benchmark run did not return any relevant results for this metric.

Binary size

This benchmark run did not return any relevant results for this metric.

Bootstrap: 481.81s -> 483.075s (0.26%)
Artifact size: 394.88 MiB -> 394.78 MiB (-0.03%)

tgross35 and others added 30 commits February 10, 2026 00:56

symcheck: Switch to getopts for argument parsing

0d2a5d2

It would be nice to support a few more flags here, but the current implementation is a bit limited. Switch to `getopts` which should make things a bit easier in the future.

symcheck: Always check args, simplify running functions

1da633c

symcheck: Clean up test setup that requires environment

7d5135c

symcheck: Don't check for empty symbol tables with PE binaries

44b0bd9

PE executables don't seem to have anything in the symbol table. Don't assert that we find any symbols in this case, which allows using symcheck for executable binaries.

Revert "ci: Temporarily disable native PPC and s390x jobs"

8f0208e

These were disabled due to permission issues, which should now be resolved. This reverts commit 28e0556a7bca6ca828186d9de74d30c7cb45db5e.

Revert "ci: Pin rustc on the native PowerPC job"

804a1f8

The nightly has the LLVM bump that resolves the relevant issue. This reverts commit 99d1fc7752b0e09cff5729cd4f3783d23c23cb66. Link: rust-lang#152428

libm: Convert frexp and ilogb to a generic implementations

f2bec36

libm: Add frexpf128, ilogbf16, and ilogbf128

212652c

The `frexpf16` signature is added (to make traits easier) but left unimplemented, since it will need a slightly different algorithm.

util: f16 now has parsing in the standard library, use it

0a4613c

util: Add support for decomposing floats

6aa1e69

util: Add an alias for the eval subcommand

c165384

libm: Print Hexf<i32> with 0x and zero padding

e7e9546

libm: Switch to a non-recursive algorithm for subnormals

4401daa

This will support `frexpf16` and should allow `generic::frexp` to be inlined into the non-generic `frexp*` functions. Additionally remove an unneeded `as` cast.

libm: Add frexpf16

67ab9a5

The new algorithm now correctly supports the type.

ci: Increase the benchmark rustc version to 2026-02-10

e68fc28

This includes the most recent LLVM bump.

libm: Improve debug output for Status

0b0d40d

libm: Add an optimization for floor

8c26adc

ceil seems to optimize better, but this gets closer to the pre-fix codegen.

libm: Add an optimization for trunc

dffcc20

Suggested-by: Juho Kahala <57393910+quaternic@users.noreply.github.com>

ci: Rewrap the command in miri.sh

873e96b

libm: Reenable should_panic tests on ppc64le

87166b5

The tests pass successfully for me on powerpc64le-unknown-linux-gnu with nightly 1.95.

libm: Reenable sincosf tests on ppc64

7d9b412

This was missed in 6e91b0346c3d (“Revert "Disable broken powerpc64 test due to https://github.com/rust-lang/rust/issues/88520"”).

ci: Allow specifying extra extensive tests to run

49e594e

For cases where we would like to run tests that aren't automatically detected, allow the following syntax in PR descriptions: ci: extra-extensive=copysignf,sqrtf16

meta: Add a root Cargo.lock file

9be3b21

In order to improve upon some inconsistencies seen across CI runs and locally, add a Cargo.lock file for the main workspace.

test: Enable the wasmbind feature on indicatif for wasm

1542d64

As of indicatif 0.18.4, this option must be enabled in able to build.

tgross35 and others added 7 commits March 6, 2026 04:18

c-b: fix path to CONTRIBUTING.md in README.md

c1d0e35

c-b: Export fminimum_num, fmaximum_num and variants

aeab5b5

Since we are starting to use the `fmaximum_num` intrinsics, LLVM may emit these.

support: Add Hexi for printing integers in hex

67633ab

Add a wrapper similar to `Hexf` to print integers with proper padding.

support: Add different NaN types to the Float trait

ccd1042

Make `Float::NAN` a qNaN, add sNaNs, and add operations to check whether a NaN is quiet or signaling. The MIPS values are checked against https://gcc.godbolt.org/z/1bhsd5ana.

support: Print sNaN or qNaN for hex floats

3a35249

Give more descriptive output in tests since we sometimes need to treat these differently. We still don't parse `sNaN`/`qNaN` for now, though we could in the future.

min,max: Add tests for signaling NaNs and update documentation

7a698be

We do handle signaling NaNs properly, with the exception of raising exceptions as IEEE 754 requires. Add tests to this effect for `fmin`, `fminimum`, `fminimum_num`, and the max variants.

tgross35 force-pushed the update-builtins branch from 35d50bd to 7a698be Compare March 14, 2026 12:45

tgross35 mentioned this pull request Mar 14, 2026

c-b: Export fminimum_num, fmaximum_num and variants rust-lang/compiler-builtins#1107

Merged

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 Mar 14, 2026

This comment has been minimized.

Sign in to view

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 Mar 14, 2026

rust-bors bot merged commit 35f1109 into rust-lang:main Mar 14, 2026
12 checks passed

rustbot added this to the 1.96.0 milestone Mar 14, 2026

tgross35 deleted the update-builtins branch March 15, 2026 09:49

RalfJung mentioned this pull request Mar 15, 2026

use correct LLVM intrinsic for min/max on floats #153343

Merged

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

Uh oh!

compiler-builtins subtree update#153866

compiler-builtins subtree update#153866
rust-bors[bot] merged 50 commits intorust-lang:mainfrom
tgross35:update-builtins

tgross35 commented Mar 14, 2026 •

edited

Loading

Uh oh!

rustbot commented Mar 14, 2026

Uh oh!

RalfJung commented Mar 14, 2026 •

edited

Loading

Uh oh!

tgross35 commented Mar 14, 2026

Uh oh!

RalfJung commented Mar 14, 2026

Uh oh!

tgross35 commented Mar 14, 2026

Uh oh!

rust-bors bot commented Mar 14, 2026

Uh oh!

This comment has been minimized.

rust-bors bot commented Mar 14, 2026

Uh oh!

Uh oh!

github-actions bot commented Mar 14, 2026

Uh oh!

rust-timer commented Mar 14, 2026

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

9 participants

Uh oh!

Conversation

tgross35 commented Mar 14, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

rustbot commented Mar 14, 2026

Uh oh!

RalfJung commented Mar 14, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

tgross35 commented Mar 14, 2026

Uh oh!

RalfJung commented Mar 14, 2026

Uh oh!

tgross35 commented Mar 14, 2026

Uh oh!

rust-bors bot commented Mar 14, 2026

Uh oh!

This comment has been minimized.

rust-bors bot commented Mar 14, 2026

Uh oh!

Uh oh!

github-actions bot commented Mar 14, 2026

Test differences

Job duration changes

Uh oh!

rust-timer commented Mar 14, 2026

Overall result: ❌✅ regressions and improvements - no action needed

Instruction count

Max RSS (memory usage)

Cycles

Binary size

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

9 participants

tgross35 commented Mar 14, 2026 •

edited

Loading

RalfJung commented Mar 14, 2026 •

edited

Loading