Skip to content

Fix 6 clippy warnings in AVX-512 microkernels #6

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
AdaWorldAPI merged 1 commit into from
Mar 16, 2026
Merged

Fix 6 clippy warnings in AVX-512 microkernels #6

Changes from all commits
Commits
Show all changes
1 commit
Select commit
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
12 changes: 6 additions & 6 deletions src/backend/kernels_avx512.rs
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -607,7 +607,7 @@ unsafe fn sgemm_ukernel_6x16(
let bv = _mm512_loadu_ps(b_packed[b_off..].as_ptr());

let a_off = p * SGEMM_MR;
c0 = _mm512_fmadd_ps(_mm512_set1_ps(a_packed[a_off + 0]), bv, c0);
c0 = _mm512_fmadd_ps(_mm512_set1_ps(a_packed[a_off]), bv, c0);
c1 = _mm512_fmadd_ps(_mm512_set1_ps(a_packed[a_off + 1]), bv, c1);
c2 = _mm512_fmadd_ps(_mm512_set1_ps(a_packed[a_off + 2]), bv, c2);
c3 = _mm512_fmadd_ps(_mm512_set1_ps(a_packed[a_off + 3]), bv, c3);
Expand All @@ -634,8 +634,8 @@ unsafe fn sgemm_ukernel_6x16(
} else {
// Masked store for edge tiles
let mask: u16 = (1u32 << nr_eff) as u16 - 1;
let cv = _mm512_maskz_loadu_ps(mask.into(), row_ptr);
_mm512_mask_storeu_ps(row_ptr, mask.into(), _mm512_add_ps(cv, rows[ir]));
let cv = _mm512_maskz_loadu_ps(mask, row_ptr);
_mm512_mask_storeu_ps(row_ptr, mask, _mm512_add_ps(cv, rows[ir]));
}
}
}
Expand Down Expand Up @@ -791,7 +791,7 @@ unsafe fn dgemm_ukernel_6x8(
let bv = _mm512_loadu_pd(b_packed[b_off..].as_ptr());

let a_off = p * DGEMM_MR;
c0 = _mm512_fmadd_pd(_mm512_set1_pd(a_packed[a_off + 0]), bv, c0);
c0 = _mm512_fmadd_pd(_mm512_set1_pd(a_packed[a_off]), bv, c0);
c1 = _mm512_fmadd_pd(_mm512_set1_pd(a_packed[a_off + 1]), bv, c1);
c2 = _mm512_fmadd_pd(_mm512_set1_pd(a_packed[a_off + 2]), bv, c2);
c3 = _mm512_fmadd_pd(_mm512_set1_pd(a_packed[a_off + 3]), bv, c3);
Expand All @@ -815,8 +815,8 @@ unsafe fn dgemm_ukernel_6x8(
_mm512_storeu_pd(row_ptr, _mm512_add_pd(cv, rows[ir]));
} else {
let mask: u8 = (1u16 << nr_eff) as u8 - 1;
let cv = _mm512_maskz_loadu_pd(mask.into(), row_ptr);
_mm512_mask_storeu_pd(row_ptr, mask.into(), _mm512_add_pd(cv, rows[ir]));
let cv = _mm512_maskz_loadu_pd(mask, row_ptr);
_mm512_mask_storeu_pd(row_ptr, mask, _mm512_add_pd(cv, rows[ir]));
}
}
}
Expand Down