Skip to content

Require fma and f16c for avx512f in std_detect #1781

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
merged 1 commit into from
May 1, 2025

Conversation

sayantn
Copy link
Contributor

@sayantn sayantn commented Apr 18, 2025

The target feature avx512f implies avx2, fma and f16c in rustc, but Intel only guarantees that avx2 will be there when avx512f is present, nothing about fma or f16c. This is correct behavior by rustc, and is based on how the assembler decides what encoding to use, VEX vs EVEX. So this PR simply requires that fma and f16c be present when enabling any AVX512 target features, to match exactly with rustc behavior.

Related:

@rustbot
Copy link
Collaborator

rustbot commented Apr 18, 2025

r? @Amanieu

rustbot has assigned @Amanieu.
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

@Amanieu Amanieu added this pull request to the merge queue May 1, 2025
Merged via the queue into rust-lang:master with commit f1c1839 May 1, 2025
58 checks passed
@sayantn sayantn deleted the fix-avx512f branch May 1, 2025 14:28
VlaDexa added a commit to VlaDexa/rust that referenced this pull request May 2, 2025
Remove `avx512dq` and `avx512vl` implication for `avx512fp16`

According to Intel, `avx512fp16` requires only `avx512bw`, but LLVM also enables `avx512vl` and `avx512dq` when `avx512fp16` is active. This is relic code, and will be fixed in LLVM soon. We should remove this from Rust too asap, especially before the stabilization of AVX512

Related:
 - llvm/llvm-project#136209
 - rust-lang#138940
 - rust-lang/stdarch#1781
 - rust-lang#111137

`@rustbot` label O-x86_64 O-x86_32 A-SIMD A-target-feature T-compiler -T-libs
r? `@Amanieu`

**Update: the LLVM fix has been merged**

cc `@rust-lang/wg-llvm` will it be possible to update the rustc llvm version to something after llvm/llvm-project#137450
Zalathar added a commit to Zalathar/rust that referenced this pull request May 2, 2025
Remove `avx512dq` and `avx512vl` implication for `avx512fp16`

According to Intel, `avx512fp16` requires only `avx512bw`, but LLVM also enables `avx512vl` and `avx512dq` when `avx512fp16` is active. This is relic code, and will be fixed in LLVM soon. We should remove this from Rust too asap, especially before the stabilization of AVX512

Related:
 - llvm/llvm-project#136209
 - rust-lang#138940
 - rust-lang/stdarch#1781
 - rust-lang#111137

``@rustbot`` label O-x86_64 O-x86_32 A-SIMD A-target-feature T-compiler -T-libs
r? ``@Amanieu``

**Update: the LLVM fix has been merged**

cc ``@rust-lang/wg-llvm`` will it be possible to update the rustc llvm version to something after llvm/llvm-project#137450
rust-timer added a commit to rust-lang-ci/rust that referenced this pull request May 2, 2025
Rollup merge of rust-lang#140389 - sayantn:avx512fp16, r=Amanieu

Remove `avx512dq` and `avx512vl` implication for `avx512fp16`

According to Intel, `avx512fp16` requires only `avx512bw`, but LLVM also enables `avx512vl` and `avx512dq` when `avx512fp16` is active. This is relic code, and will be fixed in LLVM soon. We should remove this from Rust too asap, especially before the stabilization of AVX512

Related:
 - llvm/llvm-project#136209
 - rust-lang#138940
 - rust-lang/stdarch#1781
 - rust-lang#111137

``@rustbot`` label O-x86_64 O-x86_32 A-SIMD A-target-feature T-compiler -T-libs
r? ``@Amanieu``

**Update: the LLVM fix has been merged**

cc ``@rust-lang/wg-llvm`` will it be possible to update the rustc llvm version to something after llvm/llvm-project#137450
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

Successfully merging this pull request may close these issues.

3 participants