- version update to 0.4.37
0.4.37
======
- enable neon instructions on Apple ARM64 (Aleix Conchillo Flaqué)
- orcc: Fix regression, was hard-coded to use "sse" as default target (Sebastian Dröge)
- MMX backend fixes (L. E. Segovia, Jorge Zapata)
- testsuite: Build fixes for Clang (L. E. Segovia)
- testsuite, tools: Fix warning caused by inserting unneeded source operands (L. E. Segovia)
- orccompiler: call sys_icache_invalidate() to invalidate macos inst cache (Aleix Conchillo Flaqué)
- macOS/iOS version/target check build fixes (Aleix Conchillo Flaqué)
0.4.36
======
- Only use AVX / AVX2 instructions on CPUs that support both AVX and AVX2
(fixes crash on machines that only support AVX) (L. E. Segovia)
0.4.35
======
- Add support for AVX / AVX2 (L. E. Segovia)
- SSE backend improvements (L. E. Segovia)
- New `orf` and `andf` opcodes for bitwise AND and OR for single precision floats (Jorge Zapata)
- Add support for `convwf`, int16 to float conversion (Jorge Zapata)
- Allow backend selection through ORC_TARGET environment variable (L. E. Segovia)
- Documentation improvements (Jorge Zapata, L. E. Segovia, Tim-Philipp Müller)
- orconce: Use Win32 once implementation with MSVC (Seungha Yang, L. E. Segovia)
- orcc: add --binary option to output raw machine code for functions (L. E. Segovia)
- orcprofile: Implement Windows high-resolution timestamp for MSVC
to allow benchmarking on MSVC builds (L. E. Segovia)