feat: benchmark binary dot product & min max residual

[kemingy]

No description provided.

[Copilot]

Copilot reviewed 6 out of 7 changed files in this pull request and generated 1 comment.

Files not reviewed (1)

• src/distance.rs: Evaluated as low risk

Comments suppressed due to low confidence (4)

src/kmeans.rs:132

• [nitpick] The debug message format in rabitq_assign_parallel should match the format in rabitq_assign for consistency.

"RaBitQ: rough_cmp({}), precise_cmp({}), ratio({})",

src/kmeans.rs:306

• Ensure that the new base_assign_parallel function is covered by tests to verify its behavior in multi-threaded execution.

base_assign_parallel(&vecs, &centroids, dim, self.distance, &mut labels);

src/kmeans.rs:311

• Ensure that the new rabitq_assign_parallel function is covered by tests to verify its behavior in multi-threaded execution.

rabitq_assign_parallel(&vecs, &centroids, dim, &mut labels);

src/rabitq.rs:105

• [nitpick] The function name min_max_native is ambiguous and inconsistent with the naming convention used elsewhere. It should be renamed to min_max_residual_native to be more descriptive and consistent.

pub fn min_max_native(res: &mut [f32], x: &[f32], y: &[f32]) -> (f32, f32) {

[Copilot]

Copilot reviewed 6 out of 8 changed files in this pull request and generated 2 comments.

Files not reviewed (2)

• src/distance.rs: Evaluated as low risk
• src/rabitq.rs: Evaluated as low risk

Comments suppressed due to low confidence (3)

src/simd.rs:551

• The function binary_dot_product_simd lacks test coverage. Please add test cases to ensure its correctness and handle edge cases.

pub unsafe fn binary_dot_product_simd(lhs: &[u64], rhs: &[u64]) -> u32 {

src/kmeans.rs:22

• [nitpick] The function name 'base_assign' could be more descriptive. Consider renaming it to 'assign_single_thread' for clarity.

pub fn base_assign(

src/kmeans.rs:51

• [nitpick] The function name 'base_assign_parallel' could be more descriptive. Consider renaming it to 'assign_multi_thread' for clarity.

pub fn base_assign_parallel(