Compare→bitmask SIMD lowering: where custom SIMD actually helps#8280

Draft

joseph-isaacs wants to merge 10 commits into

claude/vector-bitpack-lowering-yKOkC

Commits on Jun 6, 2026

docs: audit collect_bool/bitmask-pack sites for AVX-512 mask-compare lowering

joseph-isaacs
authored and
claude
committed
perf(vortex-buffer): SIMD fast path for packing truthy bytes into a bitmask

joseph-isaacs
authored and
claude
committed
bench(vortex-buffer): add typed between(i32) compare->bitmask benchmarks

joseph-isaacs
authored and
claude
committed
bench/docs: adjacent-pair compare bench + std::simd generalization findings

joseph-isaacs
authored and
claude
committed
docs: correct collect-bool audit per adversarial re-review

joseph-isaacs
authored and
claude
committed
feat(vortex-buffer): add safe slice-predicate bitmask packer

joseph-isaacs
authored and
claude
committed
refactor(vortex-array): route primitive/decimal `between` through safe slice packer

joseph-isaacs
authored and
claude
committed
style(vortex-buffer): rustfmt the pack_slice_predicate test attribute

joseph-isaacs
authored and
claude
committed

Commits on Jun 7, 2026

bench(vortex-buffer): scalar-vs-SIMD compare->bitmask matrix

joseph-isaacs
authored and
claude
committed
ci: drop superseded audit doc; silence bench single-char-names lint

joseph-isaacs
authored and
claude
committed