perf: SIMD auto-vectorization — HADDPD reduction + AVX2 f64×4 widening#11

Merged

jailalawat merged 3 commits into

mainfrom

issue-1-simd-reduction

May 18, 2026

Commits on May 18, 2026

perf: SIMD auto-vectorization — HADDPD reduction + AVX2 f64x4 widening (Issue #1)
jailalawat
committed
perf: AVX-512 8-wide SLP widening with runtime CPUID guard (Issue #1 Task C)
jailalawat
committed
fix: float literals lose fractional part due to jda0 shift miscompilation
jailalawat
committed