cmp_u32

simd_kernels::kernels::comparison

Function cmp_u32 

Source 
pub fn cmp_u32(
    lhs: &[u32],
    rhs: &[u32],
    mask: Option<&Bitmask>,
    op: ComparisonOperator,
) -> Result<BooleanArray<()>, KernelError>
Expand description

Type-specific SIMD-accelerated comparison function with vectorised operations.

Specialised comparison implementation optimised for the specific numeric type with architecture-appropriate lane configuration. Features memory alignment checking, SIMD vectorisation, and optional null mask support for maximum performance.

§Parameters

  • lhs: Left-hand side slice for comparison
  • rhs: Right-hand side slice for comparison
  • mask: Optional validity mask applied after comparison
  • op: Comparison operator to apply

§Returns

Result<BooleanArray<()>, KernelError> containing comparison results.

§SIMD Optimisations

  • Memory alignment: Checks 64-byte alignment for optimal SIMD operations
  • Vectorised comparisons: Uses SIMD compare operations for parallel processing
  • Scalar fallback: Efficient scalar path for unaligned or remainder elements