make abi_unsupported_vector_types a hard error #139309

RalfJung · 2025-04-03T09:05:14Z

Fixes #116558 by completing the transition; see that issue for context. The lint was introduced with Rust 1.84 and this has been shown in cargo's future breakage reports since Rust 1.85, released 6 weeks ago, and so far we got 0 complaints by users. There's not even a backlink on the tracking issue. We did a crater run when the lint was originally added and found no breakage. So I don't think we need another crater run now, but I can do one if the team prefers that.

#131800 is done, so for most current targets (in particular, all tier 1 and tier 2 targets) we have the information to implement this check (modulo the targets where we don't properly support SIMD vectors yet, see the sub-issues of #116558). If a new target gets added in the future, it will default to reject all SIMD vector types until proper information is added, which is the default we want.

This will need approval by for @rust-lang/lang. Cc @workingjubilee @veluca93

try-job: test-various
try-job: armhf-gnu
try-job: dist-i586-gnu-i586-i686-musl

rustbot · 2025-04-03T09:05:19Z

r? @fee1-dead

rustbot has assigned @fee1-dead.
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

RalfJung · 2025-04-03T11:32:18Z

tests/codegen/regparm-inreg.rs

@@ -3,7 +3,7 @@
 // x86 only.

 //@ add-core-stubs
-//@ compile-flags: --target i686-unknown-linux-gnu -Cno-prepopulate-passes -Copt-level=3
+//@ compile-flags: --target i686-unknown-linux-gnu -Cno-prepopulate-passes -Copt-level=3 -Ctarget-feature=+avx


@azhogin this test was using SIMD vector types without the appropriate target feature... I hope adding the target feature preserves the meaning of the test.

rustbot · 2025-04-03T13:45:58Z

Some changes occurred in compiler/rustc_codegen_cranelift

cc @bjorn3

compiler/rustc_codegen_cranelift/example/std_example.rs

rustbot · 2025-04-03T14:51:50Z

This PR modifies run-make tests.

cc @jieyouxu

RalfJung · 2025-04-03T17:05:34Z

tests/run-make/simd-ffi/rmake.rs

@@ -56,7 +56,7 @@ fn main() {
            .target(&target)
            .emit("llvm-ir,asm")
            .input("simd.rs")
-            .arg("-Ctarget-feature=-soft-float,+neon,+sse")
+            .arg("-Ctarget-feature=-soft-float,+neon,+sse2,+msa")
            .arg(&format!("-Cextra-filename=-{target}"))
            .run();


This test will become a complete nightmare to maintain when we start to properly enforce some basic consistency of target features.

ehuss · 2025-04-03T17:31:23Z

@RalfJung Do you think we can get this documented in the reference? Like maybe somewhere in the target_feature documentation or probably it makes more sense to be a part of the abi chapter with the abi chapter rework? Although I'm a little confused since Connor removed the SIMD section and I'm not sure why.

My uncertainty is emphasized because we currently don't really document SIMD, and I don't know how we plan to do that.

RalfJung · 2025-04-04T06:11:23Z

@RalfJung Do you think we can get this documented in the reference? Like maybe somewhere in the target_feature documentation or probably it makes more sense to be a part of the abi chapter with the abi chapter rework? Although I'm a little confused since Connor removed the SIMD section and I'm not sure why.

My uncertainty is emphasized because we currently don't really document SIMD, and I don't know how we plan to do that.

Seems like this should be part of rust-lang/reference#1545 so I'd rather not add a separate PR now, that would just be a huge conflict.

bors · 2025-04-09T06:46:15Z

☔ The latest upstream changes (presumably #139555) made this pull request unmergeable. Please resolve the merge conflicts.

bors · 2025-04-23T08:05:55Z

☀️ Try build successful - checks-actions
Build commit: 38a9dff (38a9dffa5728c708611806992f4b44be159706a8)

RalfJung · 2025-04-23T08:14:40Z

@bors r=fee1-dead,traviscross

bors · 2025-04-23T08:14:43Z

📌 Commit 64d0bb2 has been approved by fee1-dead,traviscross

It is now in the queue for this repository.

…s, r=fee1-dead,traviscross make abi_unsupported_vector_types a hard error Fixes rust-lang#116558 by completing the transition; see that issue for context. The lint was introduced with Rust 1.84 and this has been shown in cargo's future breakage reports since Rust 1.85, released 6 weeks ago, and so far we got 0 complaints by users. There's not even a backlink on the tracking issue. We did a [crater run](rust-lang#127731 (comment)) when the lint was originally added and found no breakage. So I don't think we need another crater run now, but I can do one if the team prefers that. rust-lang#131800 is done, so for most current targets (in particular, all tier 1 and tier 2 targets) we have the information to implement this check (modulo the targets where we don't properly support SIMD vectors yet, see the sub-issues of rust-lang#116558). If a new target gets added in the future, it will default to reject all SIMD vector types until proper information is added, which is the default we want. This will need approval by for `@rust-lang/lang.` Cc `@workingjubilee` `@veluca93` try-job: test-various try-job: armhf-gnu

bors · 2025-04-23T08:21:45Z

⌛ Testing commit 64d0bb2 with merge 49fa644...

bors · 2025-04-23T09:33:14Z

💔 Test failed - checks-actions

RalfJung · 2025-04-23T12:01:38Z

@bors try

…s, r=<try> make abi_unsupported_vector_types a hard error Fixes rust-lang#116558 by completing the transition; see that issue for context. The lint was introduced with Rust 1.84 and this has been shown in cargo's future breakage reports since Rust 1.85, released 6 weeks ago, and so far we got 0 complaints by users. There's not even a backlink on the tracking issue. We did a [crater run](rust-lang#127731 (comment)) when the lint was originally added and found no breakage. So I don't think we need another crater run now, but I can do one if the team prefers that. rust-lang#131800 is done, so for most current targets (in particular, all tier 1 and tier 2 targets) we have the information to implement this check (modulo the targets where we don't properly support SIMD vectors yet, see the sub-issues of rust-lang#116558). If a new target gets added in the future, it will default to reject all SIMD vector types until proper information is added, which is the default we want. This will need approval by for `@rust-lang/lang.` Cc `@workingjubilee` `@veluca93` try-job: test-various try-job: armhf-gnu try-job: dist-i586-gnu-i586-i686-musl

bors · 2025-04-23T12:02:50Z

⌛ Trying commit 216fac3 with merge d9ee062...

bors · 2025-04-23T13:37:02Z

☀️ Try build successful - checks-actions
Build commit: d9ee062 (d9ee0628138f28284b1676cc3a9fd8041f060f97)

RalfJung · 2025-04-23T13:54:28Z

@bors r=fee1-dead,traviscross

bors · 2025-04-23T13:54:31Z

📌 Commit 216fac3 has been approved by fee1-dead,traviscross

It is now in the queue for this repository.

bors · 2025-04-24T00:44:44Z

⌛ Testing commit 216fac3 with merge c02a4f0...

bors · 2025-04-24T03:52:04Z

☀️ Test successful - checks-actions
Approved by: fee1-dead,traviscross
Pushing c02a4f0 to master...

github-actions · 2025-04-24T03:53:35Z

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 fa58ce3 (parent) -> c02a4f0 (this PR)

Test differences

Show 279 test diffs

279 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 c02a4f0852e6665cf3df3867982021383f5615df --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

x86_64-apple-2: 3888.9s -> 5451.5s (40.2%)
x86_64-apple-1: 8658.0s -> 7714.6s (-10.9%)
dist-aarch64-apple: 5840.1s -> 5216.9s (-10.7%)
dist-x86_64-apple: 8480.5s -> 9242.5s (9.0%)
dist-arm-linux: 5839.8s -> 5420.9s (-7.2%)
x86_64-gnu: 6779.4s -> 6389.3s (-5.8%)
i686-gnu-2: 6419.6s -> 6739.7s (5.0%)
x86_64-gnu-llvm-20-3: 7140.1s -> 6788.2s (-4.9%)
x86_64-msvc-ext3: 7549.6s -> 7891.1s (4.5%)
dist-powerpc-linux: 5438.4s -> 5193.6s (-4.5%)

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 · 2025-04-24T05:22:51Z

Finished benchmarking commit (c02a4f0): comparison URL.

Overall result: no relevant changes - no action needed

@rustbot label: -perf-regression

Instruction count

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

Max RSS (memory usage)

Results (primary -0.4%, secondary -3.8%)

This is a less reliable metric that may be of interest but was not used to determine the overall result at the top of this comment.

	mean	range	count
Regressions ❌ (primary)	0.5%	[0.4%, 0.5%]	2
Regressions ❌ (secondary)	-	-	0
Improvements ✅ (primary)	-1.4%	[-2.2%, -0.5%]	2
Improvements ✅ (secondary)	-3.8%	[-3.8%, -3.8%]	1
All ❌✅ (primary)	-0.4%	[-2.2%, 0.5%]	4

Cycles

Results (primary 0.5%)

This is a less reliable metric that may be of interest but was not used to determine the overall result at the top of this comment.

	mean	range	count
Regressions ❌ (primary)	0.5%	[0.4%, 0.6%]	2
Regressions ❌ (secondary)	-	-	0
Improvements ✅ (primary)	-	-	0
Improvements ✅ (secondary)	-	-	0
All ❌✅ (primary)	0.5%	[0.4%, 0.6%]	2

Binary size

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

Bootstrap: 775.233s -> 774.94s (-0.04%)
Artifact size: 365.14 MiB -> 365.09 MiB (-0.01%)

rustbot assigned fee1-dead Apr 3, 2025

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. labels Apr 3, 2025

RalfJung added I-lang-nominated Nominated for discussion during a lang team meeting. I-lang-easy-decision Issue: The decision needed by the team is conjectured to be easy; this does not imply nomination labels Apr 3, 2025