rust-lang/rust
1
0
Fork 0
Code Issues Pull Requests Actions 1 Packages Projects Releases Wiki Activity

fix: reversed _mm512_set4_epi64 implementation

Browse Source 
Fixes #1555 by changing the implementations of _mm512_set4_epi64 and _mm512_setr4_epi64 to use _mm512_set_epi64. This makes these implementations consistent with the other _mm512_set[r]4_* implementations as well as changes their behavior to be in line with what the intrinsics guide describes.
This commit is contained in:
Kirsten Laskoski
2024-04-07 18:16:31 -04:00
committed by Amanieu d'Antras
parent 06495b6f45
commit 39ab14c01b
1 changed files with 2 additions and 4 deletions
Download Patch File Download Diff File Expand all files Collapse all files

6
library/stdarch/crates/core_arch/src/x86/avx512f.rs
View File

@@ -28363,8 +28363,7 @@ pub unsafe fn _mm_maskz_set1_epi64(k: __mmask8, a: i64) -> __m128i {
#[target_feature(enable = "avx512f")] #[target_feature(enable = "avx512f")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")] #[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
pub unsafe fn _mm512_set4_epi64(d: i64, c: i64, b: i64, a: i64) -> __m512i { pub unsafe fn _mm512_set4_epi64(d: i64, c: i64, b: i64, a: i64) -> __m512i {
let r = i64x8::new(d, c, b, a, d, c, b, a); _mm512_set_epi64(d, c, b, a, d, c, b, a)
transmute(r)
} }
/// Set packed 64-bit integers in dst with the repeated 4 element sequence in reverse order. /// Set packed 64-bit integers in dst with the repeated 4 element sequence in reverse order.
@@ -28374,8 +28373,7 @@ pub unsafe fn _mm512_set4_epi64(d: i64, c: i64, b: i64, a: i64) -> __m512i {
#[target_feature(enable = "avx512f")] #[target_feature(enable = "avx512f")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")] #[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
pub unsafe fn _mm512_setr4_epi64(d: i64, c: i64, b: i64, a: i64) -> __m512i { pub unsafe fn _mm512_setr4_epi64(d: i64, c: i64, b: i64, a: i64) -> __m512i {
let r = i64x8::new(a, b, c, d, a, b, c, d); _mm512_set_epi64(a, b, c, d, a, b, c, d)
transmute(r)
} }
/// Compare packed single-precision (32-bit) floating-point elements in a and b for less-than, and store the results in mask vector k. /// Compare packed single-precision (32-bit) floating-point elements in a and b for less-than, and store the results in mask vector k.