Fix is_ascii performance regression on AVX-512 CPUs when compiling with -C target-cpu=native

[bonega]

Summary

Fix [u8]::is_ascii() performance regression when compiled with -C target-cpu=native on AVX-512 CPUs.

Problem

When is_ascii is compiled with AVX-512 enabled, LLVM's auto-vectorization generates ~31 kshiftrd instructions to extract mask bits one-by-one, instead of using the efficient pmovmskb
instruction. This causes a ~22x performance regression.

Because is_ascii is marked #[inline], it gets inlined and recompiled with the user's target settings, affecting anyone using -C target-cpu=native on AVX-512 CPUs.

Solution

Replace the counting loop with explicit SSE2 intrinsics (_mm_movemask_epi8) that force pmovmskb codegen regardless of CPU features.

Godbolt Links (Rust 1.92)

Pattern	Target	Link	Result
Counting loop (old)	Default SSE2	https://godbolt.org/z/sE86xz4fY	pmovmskb
Counting loop (old)	AVX-512 (znver4)	https://godbolt.org/z/b3jvMhGd3	31x kshiftrd (broken)
SSE2 intrinsics (fix)	Default SSE2	https://godbolt.org/z/hMeGfeaPv	pmovmskb
SSE2 intrinsics (fix)	AVX-512 (znver4)	https://godbolt.org/z/Tdvdqjohn	vpmovmskb (fixed)

Benchmark Results

AMD Ryzen 5 7500F (Zen 4 with AVX-512):

Build	Before	After	Improvement
Default	~73 GB/s	~74 GB/s	No regression
-C target-cpu=native	~3 GB/s	~67 GB/s	22x

Note: this is the pure ascii path, but the story is similar for the others.
See linked bench project.

Test Plan

• Assembly test (slice-is-ascii-avx512.rs) verifies no kshiftrd with AVX-512
• Existing codegen test updated to loongarch64-only (auto-vectorization still used there)
• Fuzz testing confirms old/new implementations produce identical results (~53M iterations)
• Benchmarks confirm performance improvement
• Tidy checks pass

Reproduction / Test Projects

Standalone validation tools: https://github.com/bonega/is-ascii-fix-validation

• bench/ - Criterion benchmarks for SSE2 vs AVX-512 comparison
• fuzz/ - Compares old/new implementations with libfuzzer

Related Issues

• Regression introduced in Optimize is_ascii for str and [u8] further #130733
• Similar AVX-512 kshiftrd codegen issue: Weird AVX 512 code generated with std::simd when using -Zbuild-std  #129293

[rustbot]

⚠️ #[rustc_allow_const_fn_unstable] needs careful audit to avoid accidentally exposing unstable
implementation details on stable.

cc @rust-lang/wg-const-eval

[rustbot]

r? @tgross35

rustbot has assigned @tgross35.
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