Implement "real" AVX2 intrinsics and clean up x86 codegen - #115

Merged

valadaptive merged 11 commits into

linebender:mainfrom

valadaptive:more-avx2

Nov 14, 2025

Commits on Nov 12, 2025

Improve test coverage a bit
valadaptive
committed
21967e9
Implement AVX2 and improve x86 codegen
valadaptive
committed
210a712
Consolidate the x86 codegen
valadaptive
committed
97a0494
Add more split/combine tests
valadaptive
committed
c4f15a1
Add specialized AVX2 split/combine ops
valadaptive
committed
0ca9b62
Add 256-bit widen/narrow tests
valadaptive
committed
1fee324
Add specialized AVX2 widen/narrow ops
valadaptive
committed
bd31e47
Fix spontaneous Clippy complaint
valadaptive
committed
843fc66
Implement 8-bit multiplication in x86
valadaptive
committed
d0bff93
Fix meaning of acceptable_wide_op in SSE4.2 gen
valadaptive
committed
b29bf2a

Commits on Nov 14, 2025

Add CHANGELOG entries
valadaptive
committed
2e4235d