diff options
| author | Scott McMurray <scottmcm@users.noreply.github.com> | 2017-06-20 23:48:15 -0700 |
|---|---|---|
| committer | Scott McMurray <scottmcm@users.noreply.github.com> | 2017-06-21 22:58:45 -0700 |
| commit | 47fa016193a729091aef9c531df9385548ad46ab (patch) | |
| tree | 9c2fdcbe55503be8d89c553a807ecc215cf2e567 /src/test/run-pass/thinlto | |
| parent | 6de26f42deab53b1a8cbd57e4c696626e6650708 (diff) | |
| download | rust-47fa016193a729091aef9c531df9385548ad46ab.tar.gz rust-47fa016193a729091aef9c531df9385548ad46ab.zip | |
Reuse the mem::swap optimizations to speed up slice::rotate
Exposes the swapping logic from PR 40454 as `pub unsafe fn ptr::swap_nonoverlapping` under feature swap_nonoverlapping This is most helpful for compound types where LLVM didn't vectorize the loop. Highlight: bench slice::rotate_medium_by727_strings gets 38% faster.
Diffstat (limited to 'src/test/run-pass/thinlto')
0 files changed, 0 insertions, 0 deletions