Expand description
SIMD-accelerated operations
This module provides SIMD-optimized parallel operations for maximum performance on modern CPUs. Falls back to scalar operations when SIMD is not available.
Functionsยง
- parallel_
simd_ sum_ f32 - Parallel SIMD sum for f32
- parallel_
simd_ sum_ f64 - Parallel SIMD sum for f64
- parallel_
simd_ sum_ i32 - Parallel SIMD sum using work distribution
- simd_
dot_ f32 - SIMD-accelerated dot product for f32
- simd_
dot_ f64 - SIMD-accelerated dot product for f64
- simd_
sum_ f32 - SIMD-accelerated sum for f32
- simd_
sum_ f64 - SIMD-accelerated sum for f64
- simd_
sum_ i32 - SIMD-accelerated sum for i32