Skip to content

chore: simplify dot implementation to use auto-vectorization #2645

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.

Sign up for GitHub

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
wjones127 merged 2 commits into from
Nov 10, 2025
Merged

chore: simplify dot implementation to use auto-vectorization #2645

Changes from all commits
Commits
Show all changes
2 commits
Select commit Hold shift + click to select a range
4efe649
simplify dot implemetation
eddyxu Jul 26, 2024
3b0a32e
fix duration import
eddyxu Jul 30, 2024
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
55 changes: 1 addition & 54 deletions rust/lance-linalg/src/distance/dot.rs
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -19,10 +19,6 @@ use lance_core::utils::cpu::SimdSupport;
use lance_core::utils::cpu::SIMD_SUPPORT;
use num_traits::{real::Real, AsPrimitive, Num};

use crate::simd::{
f32::{f32x16, f32x8},
SIMD,
};
use crate::Result;

/// Default implementation of dot product.
Expand Down Expand Up @@ -142,56 +138,7 @@ impl Dot for f16 {
impl Dot for f32 {
#[inline]
fn dot(x: &[Self], y: &[Self]) -> f32 {
// Manually unrolled 8 times to get enough registers.
// TODO: avx512 can unroll more
let x_unrolled_chunks = x.chunks_exact(64);
let y_unrolled_chunks = y.chunks_exact(64);

// 8 float32 SIMD
let x_aligned_chunks = x_unrolled_chunks.remainder().chunks_exact(8);
let y_aligned_chunks = y_unrolled_chunks.remainder().chunks_exact(8);

let sum = if x_aligned_chunks.remainder().is_empty() {
0.0
} else {
debug_assert_eq!(
x_aligned_chunks.remainder().len(),
y_aligned_chunks.remainder().len()
);
x_aligned_chunks
.remainder()
.iter()
.zip(y_aligned_chunks.remainder().iter())
.map(|(&x, &y)| x * y)
.sum()
};

let mut sum8 = f32x8::zeros();
x_aligned_chunks
.zip(y_aligned_chunks)
.for_each(|(x_chunk, y_chunk)| unsafe {
let x1 = f32x8::load_unaligned(x_chunk.as_ptr());
let y1 = f32x8::load_unaligned(y_chunk.as_ptr());
sum8 += x1 * y1;
});

let mut sum16 = f32x16::zeros();
x_unrolled_chunks
.zip(y_unrolled_chunks)
.for_each(|(x, y)| unsafe {
let x1 = f32x16::load_unaligned(x.as_ptr());
let x2 = f32x16::load_unaligned(x.as_ptr().add(16));
let x3 = f32x16::load_unaligned(x.as_ptr().add(32));
let x4 = f32x16::load_unaligned(x.as_ptr().add(48));

let y1 = f32x16::load_unaligned(y.as_ptr());
let y2 = f32x16::load_unaligned(y.as_ptr().add(16));
let y3 = f32x16::load_unaligned(y.as_ptr().add(32));
let y4 = f32x16::load_unaligned(y.as_ptr().add(48));

sum16 += (x1 * y1 + x2 * y2) + (x3 * y3 + x4 * y4);
});
sum16.reduce_sum() + sum8.reduce_sum() + sum
dot_scalar::<Self, Self, 16>(x, y)
}
}

Expand Down
Loading