Stabilize the avx512 target features

[sayantn]

This PR stabilizes the AVX512 target features - see this comment.

Tracking Issue - #44839

The target feature UI tests have been changed to x87 (chosen because this is very unlikely to stablize ever, please comment if some other feature will be better)

related: #111137

[rustbot]

r? @davidtwco

rustbot has assigned @davidtwco.
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

[rustbot]

⚠️ Warning ⚠️

• These commits modify submodules.

[rustbot]

The Miri subtree was changed

cc @rust-lang/miri

[Amanieu]

Nominating for T-lang FCP. This stabilizes the AVX512 target features which is a per-requisite for stabilizing the AVX-512 intrinsics tracked in #111137.

[ehuss]

Can you also send a PR to update https://github.com/rust-lang/reference/blob/master/src/attributes/codegen.md#x86-or-x86_64?

[Amanieu]

These features are for the various AVX512 sub-features. We have already stabilized some of these names in is_x86_feature_detected! so the language level name should match.

[rustbot]

⚠️ Warning ⚠️

• Some commits in this PR modify submodules.

[rustbot]

Reminder, once the PR becomes ready for a review, use @rustbot ready.

[sayantn]

@traviscross @Amanieu is the feature detection part of T-lang or T-libs-api.

• If it is part of T-lang, then Stabilize runtime detection of VEX variants of avx512 stdarch#1760 is good to merge now, and then this PR will be good to go with an stdarch update.
• If it is part of T-libs-api, then the fcp on Tracking Issue for AVX512 intrinsics #111137 will need to complete before merging the stdarch PR

[Amanieu]

rust-lang/stdarch#1760 can land as soon as this FCP is complete. Then you can update the stdarch submodule in this PR and it can itself be merged.

[sayantn]

@Amanieu the FCP on this PR is complete, so stdarch#1760 can be merged now, right?

[sayantn]

@traviscross finally this PR is ready for merge, just waiting for CI

[traviscross]

Thanks for pushing this along.

@bors r=Amanieu,traviscross rollup

[bors]

📌 Commit 2898680 has been approved by Amanieu,traviscross

It is now in the queue for this repository.

[Kobzol]

@rust-timer build 327d736

[sayantn]

@traviscross do I need to do anything for the relnotes PR?

[traviscross]

do I need to do anything for the relnotes PR?

Good question. If you think it could warrant a brief section in the release blog post, it'd be worth writing up something in a comment there for the release team. Otherwise, there's not anything I see to do on this one.

[rust-timer]

Finished benchmarking commit (327d736): comparison URL.

Overall result: ❌✅ regressions and improvements - please read the text below

Benchmarking this pull request likely means that it is perf-sensitive, so we're automatically marking it as not fit for rolling up. While you can manually mark this PR as fit for rollup, we strongly recommend not doing so since this PR may lead to changes in compiler perf.

Next Steps: If you can justify the regressions found in this try perf run, please indicate this with @rustbot label: +perf-regression-triaged along with sufficient written justification. If you cannot justify the regressions please fix the regressions and do another perf run. If the next run shows neutral or positive results, the label will be automatically removed.

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

Instruction count

This is the most reliable metric that we have; it was used to determine the overall result at the top of this comment. However, even this metric can sometimes exhibit noise.

	mean	range	count
Regressions ❌ (primary)	2.0%	[1.1%, 3.0%]	2
Regressions ❌ (secondary)	-	-	0
Improvements ✅ (primary)	-	-	0
Improvements ✅ (secondary)	-1.0%	[-1.0%, -1.0%]	3
All ❌✅ (primary)	2.0%	[1.1%, 3.0%]	2

Max RSS (memory usage)

Results (primary -0.5%, secondary -2.4%)

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
Regressions ❌ (secondary)	-	-	0
Improvements ✅ (primary)	-0.5%	[-0.5%, -0.5%]	1
Improvements ✅ (secondary)	-2.4%	[-5.3%, -0.4%]	6
All ❌✅ (primary)	-0.5%	[-0.5%, -0.5%]	1

Cycles

Results (primary 3.0%, secondary -3.0%)

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)	3.0%	[3.0%, 3.0%]	1
Regressions ❌ (secondary)	-	-	0
Improvements ✅ (primary)	-	-	0
Improvements ✅ (secondary)	-3.0%	[-5.1%, -0.5%]	6
All ❌✅ (primary)	3.0%	[3.0%, 3.0%]	1

Binary size

Results (primary 1.1%)

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)	1.1%	[1.1%, 1.1%]	1
Regressions ❌ (secondary)	-	-	0
Improvements ✅ (primary)	-	-	0
Improvements ✅ (secondary)	-	-	0
All ❌✅ (primary)	1.1%	[1.1%, 1.1%]	1

Bootstrap: 776.543s -> 777.319s (0.10%)
Artifact size: 365.45 MiB -> 365.33 MiB (-0.03%)

[sayantn]

@traviscross I definitely think that this would deserve a place in the relnotes, but only after the stabilization of intrinsics. I don't think the stabilization of target features by themselves are too important, but the intrinsics have been a major demand from user for quite some time.