Andrew Brown 54b45d28a3 x64: lower fcvt_from_uint to VCVTUDQ2PS when possible ...

When AVX512VL and AVX512F are available, use a single instruction
(`VCVTUDQ2PS`) instead of a length 9-instruction sequence. This
optimization is a port from the legacy x86 backend.

2021-05-19 12:20:11 -07:00

..

filetests

x64: lower fcvt_from_uint to VCVTUDQ2PS when possible

2021-05-19 12:20:11 -07:00

src

Factor out byteorder in cranelift

2021-04-23 12:05:18 -07:00

Cargo.toml

Upgrade to the latest versions of gimli, addr2line, object (#2901)

2021-05-12 10:53:17 -05:00

LICENSE

moved crates in lib/ to src/, renamed crates, modified some files' text (#660)

2019-01-28 15:56:54 -08:00