ci: wire up CI, fix no_std build, fix stale package names in scripts

- Add .gitea/workflows/ci.yml running scripts/ci-test.sh (fmt, clippy, test, no_std check) on push/PR to main. - Fix stale rustyhdf5-py/rustyhdf5-format package names in ci-test.sh/check-nostd.sh, which had been silently no-op'ing those checks (cargo warns but doesn't fail on an unknown --exclude/-p target). - With those checks actually running, fix the real issues they surface: - clippy: useless_conversion in chunked_write.rs, byte_char_slices in global_heap.rs/object_header.rs. - cargo fmt: apply formatting across the workspace (whitespace only). - no_std (thumbv7em-none-eabihf) build errors in clawhdf5-format: core::sync::atomic::AtomicU64 doesn't exist on that target (no native 64-bit atomics) — switch profiling.rs's counters to portable-atomic, which falls back to a CAS-based emulation there and is a no-op wrapper elsewhere. Add missing alloc imports for Box (filters.rs), Vec (filters_szip.rs), and format! (dict_encoding.rs) on no_std paths. Replace f64::powi (std/libm-only) with a small local exponentiation-by-squaring helper in the scale-offset filter.
2026-08-05 10:50:13 -07:00
parent b70d594c4f
commit 55959b4920
34 changed files with 590 additions and 365 deletions
@@ -13,42 +13,44 @@ use std::arch::x86_64::*;
 /// Caller must verify is_x86_feature_detected!("avx512f").
 // SAFETY: Caller must have verified avx512f via is_x86_feature_detected!.
 #[target_feature(enable = "avx512f")]
-pub unsafe fn dot_product(a: &[f32], b: &[f32]) -> f32 { unsafe {
-    assert_eq!(a.len(), b.len());
-    let len = a.len();
-    let mut i = 0;
-    let mut acc0 = _mm512_setzero_ps();
-    let mut acc1 = _mm512_setzero_ps();
+pub unsafe fn dot_product(a: &[f32], b: &[f32]) -> f32 {
+    unsafe {
+        assert_eq!(a.len(), b.len());
+        let len = a.len();
+        let mut i = 0;
+        let mut acc0 = _mm512_setzero_ps();
+        let mut acc1 = _mm512_setzero_ps();

-    // Process 32 elements per iteration (2x16 unrolled)
-    while i + 32 <= len {
-        let va0 = _mm512_loadu_ps(a.as_ptr().add(i));
-        let vb0 = _mm512_loadu_ps(b.as_ptr().add(i));
-        acc0 = _mm512_fmadd_ps(va0, vb0, acc0);
+        // Process 32 elements per iteration (2x16 unrolled)
+        while i + 32 <= len {
+            let va0 = _mm512_loadu_ps(a.as_ptr().add(i));
+            let vb0 = _mm512_loadu_ps(b.as_ptr().add(i));
+            acc0 = _mm512_fmadd_ps(va0, vb0, acc0);

-        let va1 = _mm512_loadu_ps(a.as_ptr().add(i + 16));
-        let vb1 = _mm512_loadu_ps(b.as_ptr().add(i + 16));
-        acc1 = _mm512_fmadd_ps(va1, vb1, acc1);
+            let va1 = _mm512_loadu_ps(a.as_ptr().add(i + 16));
+            let vb1 = _mm512_loadu_ps(b.as_ptr().add(i + 16));
+            acc1 = _mm512_fmadd_ps(va1, vb1, acc1);

-        i += 32;
+            i += 32;
+        }
+
+        if i + 16 <= len {
+            let va = _mm512_loadu_ps(a.as_ptr().add(i));
+            let vb = _mm512_loadu_ps(b.as_ptr().add(i));
+            acc0 = _mm512_fmadd_ps(va, vb, acc0);
+            i += 16;
+        }
+
+        let mut sum = _mm512_reduce_add_ps(_mm512_add_ps(acc0, acc1));
+
+        while i < len {
+            sum += a[i] * b[i];
+            i += 1;
+        }
+
+        sum
    }
-
-    if i + 16 <= len {
-        let va = _mm512_loadu_ps(a.as_ptr().add(i));
-        let vb = _mm512_loadu_ps(b.as_ptr().add(i));
-        acc0 = _mm512_fmadd_ps(va, vb, acc0);
-        i += 16;
-    }
-
-    let mut sum = _mm512_reduce_add_ps(_mm512_add_ps(acc0, acc1));
-
-    while i < len {
-        sum += a[i] * b[i];
-        i += 1;
-    }
-
-    sum
-}}
+}

 /// AVX-512 cosine similarity — fused single pass.
 ///
@@ -56,38 +58,40 @@ pub unsafe fn dot_product(a: &[f32], b: &[f32]) -> f32 { unsafe {
 /// Caller must verify is_x86_feature_detected!("avx512f").
 // SAFETY: Caller must have verified avx512f via is_x86_feature_detected!.
 #[target_feature(enable = "avx512f")]
-pub unsafe fn cosine_similarity(a: &[f32], b: &[f32]) -> f32 { unsafe {
-    assert_eq!(a.len(), b.len());
-    let len = a.len();
-    let mut i = 0;
+pub unsafe fn cosine_similarity(a: &[f32], b: &[f32]) -> f32 {
+    unsafe {
+        assert_eq!(a.len(), b.len());
+        let len = a.len();
+        let mut i = 0;

-    let mut dot_acc = _mm512_setzero_ps();
-    let mut norm_a_acc = _mm512_setzero_ps();
-    let mut norm_b_acc = _mm512_setzero_ps();
+        let mut dot_acc = _mm512_setzero_ps();
+        let mut norm_a_acc = _mm512_setzero_ps();
+        let mut norm_b_acc = _mm512_setzero_ps();

-    while i + 16 <= len {
-        let va = _mm512_loadu_ps(a.as_ptr().add(i));
-        let vb = _mm512_loadu_ps(b.as_ptr().add(i));
-        dot_acc = _mm512_fmadd_ps(va, vb, dot_acc);
-        norm_a_acc = _mm512_fmadd_ps(va, va, norm_a_acc);
-        norm_b_acc = _mm512_fmadd_ps(vb, vb, norm_b_acc);
-        i += 16;
+        while i + 16 <= len {
+            let va = _mm512_loadu_ps(a.as_ptr().add(i));
+            let vb = _mm512_loadu_ps(b.as_ptr().add(i));
+            dot_acc = _mm512_fmadd_ps(va, vb, dot_acc);
+            norm_a_acc = _mm512_fmadd_ps(va, va, norm_a_acc);
+            norm_b_acc = _mm512_fmadd_ps(vb, vb, norm_b_acc);
+            i += 16;
+        }
+
+        let mut dot = _mm512_reduce_add_ps(dot_acc);
+        let mut norm_a = _mm512_reduce_add_ps(norm_a_acc);
+        let mut norm_b = _mm512_reduce_add_ps(norm_b_acc);
+
+        while i < len {
+            dot += a[i] * b[i];
+            norm_a += a[i] * a[i];
+            norm_b += b[i] * b[i];
+            i += 1;
+        }
+
+        let denom = (norm_a * norm_b).sqrt();
+        if denom == 0.0 { 0.0 } else { dot / denom }
    }
-
-    let mut dot = _mm512_reduce_add_ps(dot_acc);
-    let mut norm_a = _mm512_reduce_add_ps(norm_a_acc);
-    let mut norm_b = _mm512_reduce_add_ps(norm_b_acc);
-
-    while i < len {
-        dot += a[i] * b[i];
-        norm_a += a[i] * a[i];
-        norm_b += b[i] * b[i];
-        i += 1;
-    }
-
-    let denom = (norm_a * norm_b).sqrt();
-    if denom == 0.0 { 0.0 } else { dot / denom }
-}}
+}

 /// AVX-512 L2 distance.
 ///
@@ -95,27 +99,29 @@ pub unsafe fn cosine_similarity(a: &[f32], b: &[f32]) -> f32 { unsafe {
 /// Caller must verify is_x86_feature_detected!("avx512f").
 // SAFETY: Caller must have verified avx512f via is_x86_feature_detected!.
 #[target_feature(enable = "avx512f")]
-pub unsafe fn l2_distance(a: &[f32], b: &[f32]) -> f32 { unsafe {
-    assert_eq!(a.len(), b.len());
-    let len = a.len();
-    let mut i = 0;
-    let mut acc = _mm512_setzero_ps();
+pub unsafe fn l2_distance(a: &[f32], b: &[f32]) -> f32 {
+    unsafe {
+        assert_eq!(a.len(), b.len());
+        let len = a.len();
+        let mut i = 0;
+        let mut acc = _mm512_setzero_ps();

-    while i + 16 <= len {
-        let va = _mm512_loadu_ps(a.as_ptr().add(i));
-        let vb = _mm512_loadu_ps(b.as_ptr().add(i));
-        let diff = _mm512_sub_ps(va, vb);
-        acc = _mm512_fmadd_ps(diff, diff, acc);
-        i += 16;
+        while i + 16 <= len {
+            let va = _mm512_loadu_ps(a.as_ptr().add(i));
+            let vb = _mm512_loadu_ps(b.as_ptr().add(i));
+            let diff = _mm512_sub_ps(va, vb);
+            acc = _mm512_fmadd_ps(diff, diff, acc);
+            i += 16;
+        }
+
+        let mut sum = _mm512_reduce_add_ps(acc);
+
+        while i < len {
+            let d = a[i] - b[i];
+            sum += d * d;
+            i += 1;
+        }
+
+        sum.sqrt()
    }
-
-    let mut sum = _mm512_reduce_add_ps(acc);
-
-    while i < len {
-        let d = a[i] - b[i];
-        sum += d * d;
-        i += 1;
-    }
-
-    sum.sqrt()
-}}
+}