File: reduce-fadd-unordered.rs

package info (click to toggle)
rustc 1.85.0%2Bdfsg3-1
  • links: PTS, VCS
  • area: main
  • in suites: experimental, forky, sid, trixie
  • size: 893,396 kB
  • sloc: xml: 158,127; python: 35,830; javascript: 19,497; cpp: 19,002; sh: 17,245; ansic: 13,127; asm: 4,376; makefile: 1,051; perl: 29; lisp: 29; ruby: 19; sql: 11
file content (30 lines) | stat: -rw-r--r-- 1,035 bytes parent folder | download | duplicates (3)
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
//@ revisions: x86_64 aarch64
//@ assembly-output: emit-asm
//@ compile-flags: --crate-type=lib -O
//@[aarch64] only-aarch64
//@[x86_64] only-x86_64
//@[x86_64] compile-flags: -Ctarget-feature=+sse3
//@ ignore-sgx Test incompatible with LVI mitigations
#![feature(portable_simd)]
#![feature(core_intrinsics)]
use std::intrinsics::simd as intrinsics;
use std::simd::*;
// Regression test for https://github.com/rust-lang/rust/issues/130028
// This intrinsic produces much worse code if you use +0.0 instead of -0.0 because
// +0.0 isn't as easy to algebraically reassociate, even using LLVM's reassoc attribute!
// It would emit about an extra fadd, depending on the architecture.

// CHECK-LABEL: reduce_fadd_negative_zero
pub unsafe fn reduce_fadd_negative_zero(v: f32x4) -> f32 {
    // x86_64: addps
    // x86_64-NEXT: movshdup
    // x86_64-NEXT: addss
    // x86_64-NOT: xorps

    // aarch64: faddp
    // aarch64-NEXT: faddp

    // CHECK-NOT: {{f?}}add{{p?s*}}
    // CHECK: ret
    intrinsics::simd_reduce_add_unordered(v)
}