Last active
May 13, 2024 10:49
-
-
Save 00xc/e6b160970e242c7c3f439f1ae98287ad to your computer and use it in GitHub Desktop.
Finding the lowest set bit in a 512-bit array using SIMD
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
#![feature(portable_simd)] | |
#![feature(stdsimd)] | |
use core::arch::x86_64::*; | |
unsafe fn _mm512_tzcnt_si512(v: __m512i) -> Option<usize> { | |
let zeros = _mm512_setzero_si512(); | |
let mask = _mm512_cmpneq_epi64_mask(v, zeros); | |
// 63 - lzcnt((v - 1) ^ v); | |
let unweighted = _mm512_sub_epi64( | |
_mm512_set1_epi64(63), | |
// lzcnt((v - 1) ^ v) | |
_mm512_lzcnt_epi64( | |
// (v - 1) ^ v | |
_mm512_xor_epi64( | |
v, | |
// v - 1 | |
_mm512_sub_epi64( | |
v, | |
_mm512_set1_epi64(1), | |
), | |
), | |
) | |
); | |
let weights = _mm512_set_epi64( | |
64 * 7, | |
64 * 6, | |
64 * 5, | |
64 * 4, | |
64 * 3, | |
64 * 2, | |
64 * 1, | |
64 * 0, | |
); | |
let weighted = _mm512_mask_add_epi64( | |
_mm512_set1_epi64(i64::MAX), | |
mask, | |
weights, | |
unweighted, | |
); | |
let result = _mm512_reduce_min_epi64(weighted); | |
(result < 512).then_some(result as usize) | |
} |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment