[libcxx-commits] [libcxx] [libc++] Optimize vector::__move_range for trivially copyable types (PR #207384)

[Louis Dionne via libcxx-commits]

ldionne wrote:

The benchmark results seem to be overall good on macOS (mostly neutrals and a few improvements that don't seem to be in the noise), however they seem to be overall worse on Linux. I'd like to understand why that is.

For now, let's run the benchmarks again exactly the same way. If the above is in the noise, then we shouldn't see exactly the same benchmarks display a similar swing.

/libcxx-bot benchmark libcxx/test/benchmarks/containers/sequence/vector.bench.cpp

https://github.com/llvm/llvm-project/pull/207384