Remove misleading target feature aliases

[calebzulawski]

Fixes #100752. This is a follow up to #103750. These aliases could not be completely removed until rust-lang/stdarch#1355 landed.

cc @Amanieu

[rustbot]

r? @davidtwco

(rustbot has picked a reviewer for you, use r? to override)

[Amanieu]

LGTM, but I believe this needs T-lang signoff since it affects target features. However I do believe these features are still unstable so it should be fine.

[calebzulawski]

Yep, they are still unstable.
r? @rust-lang/lang

[workingjubilee]

For context for T-lang, since it might be a bit cryptic from the remarks elsewhere:

There are in fact "avx512{gfni, vaes, ...}" instructions and so on, but the x86 CPU denotes that these instructions are available by signaling "both the avx512 bit and the feature bit enabled" when CPUID is queried. For each feature, it signals that "only" the "feature" instructions are available via having the feature bit enabled but the avx512 bit disabled. The exact details of how this works out sometimes varies a little from what I just said since x86 is a very... "evolved, not designed" instruction set architecture.

[anden3]

Hi @calebzulawski! What's the status on this PR? I saw you got a review on the 6th of February :)

[Amanieu]

This is waiting on T-lang sign-off. Otherwise the PR itself looks fine.

[nikomatsakis]

If you want lang team signoff, we need to nominate it for lang-team discussion! See instructions here:

https://lang-team.rust-lang.org/how_to/nominate.html

In particular, can you author a relatively standalone paragraph explaining the question to be answered? Thanks!

[calebzulawski]

@rustbot label +I-lang-nominated

When AVX512 support was initially added, these three features (GFNI, VAES, and VPCLMULQDQ) were included as AVX512 features and prefixed with the avx512 name. These features are not yet stabilized.

These features do provide AVX512 instructions when paired with the avx512f feature, but on their own they do not. This is not just theoretical--some Intel CPUs support e.g. GFNI that only operates on SSE and AVX vectors (without any AVX512 support). Intel (and LLVM) do not include AVX512 in the names of these features, either. Having the avx512 prefix may mislead a user into believing the avx512f feature is enabled by these features, which is true for all other AVX512 features.

My previous PR added the more accurate gfni, vaes, and vpclmulqdq target feature names. This PR introduces a nightly/unstable breaking change by removing the avx512gfni, avx512vaes, and avx512vpclmulqdq aliases for these features.

[nikomatsakis]

@rfcbot fcp merge

Discussed in @rust-lang/lang meeting. We feel good about this change, particularly since target feature gives you warnings if things are wrong.

[nikomatsakis]

@rfcbot fcp cancel

[rfcbot]

@nikomatsakis proposal cancelled.

[nikomatsakis]

@rfcbot fcp merge

Discussed in https://github.com/orgs/rust-lang/teams/lang meeting. We feel good about this change, particularly since target feature gives you warnings if things are wrong.

[rfcbot]

Team member @nikomatsakis has proposed to merge this. The next step is review by the rest of the tagged team members:

• @joshtriplett
• @nikomatsakis
• @pnkfelix
• @scottmcm
• @tmandry

No concerns currently listed.

Once a majority of reviewers approve (and at most 2 approvals are outstanding), this will enter its final comment period. If you spot a major issue that hasn't been raised at any point in this process, please speak up!

cc @rust-lang/lang-advisors: FCP proposed for lang, please feel free to register concerns.
See this document for info about what commands tagged team members can give me.

[rfcbot]

🔔 This is now entering its final comment period, as per the review above. 🔔

[nikomatsakis]

@rustbot labels -S-waiting-on-team

[nikomatsakis]

@rustbot labels -I-lang-nominated

[rfcbot]

The final comment period, with a disposition to merge, as per the review above, is now complete.

As the automated representative of the governance process, I would like to thank the author for their work and everyone else who contributed.

This will be merged soon.

[Amanieu]

@bors r+

[bors]

📌 Commit 47fc132 has been approved by Amanieu

It is now in the queue for this repository.

[bors]

⌛ Testing commit 47fc132 with merge ce5919f...

[bors]

☀️ Test successful - checks-actions
Approved by: Amanieu
Pushing ce5919f to master...

[rust-timer]

Finished benchmarking commit (ce5919f): comparison URL.

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

@rustbot label: -perf-regression

Instruction count

This is a highly reliable metric that was used to determine the overall result at the top of this comment.

	mean	range	count
Regressions ❌ (primary)	-	-	0
Regressions ❌ (secondary)	1.1%	[1.1%, 1.1%]	1
Improvements ✅ (primary)	-	-	0
Improvements ✅ (secondary)	-1.9%	[-1.9%, -1.9%]	1
All ❌✅ (primary)	-	-	0

Max RSS (memory usage)

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

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: 644.425s -> 644.092s (-0.05%)