There are some LLVM optimization talks that show otherwise, where longer vectori... | Hacker News

Hacker News new | past | comments | ask | show | jobs | submit

login

pjmlp on April 15, 2020 | parent | context | favorite | on: KolibriOS – operating system written entirely in a...

There are some LLVM optimization talks that show otherwise, where longer vectorized code actually runs faster than the smaller version, though.

userbinator on April 15, 2020 [–]

...in microbenchmarks.

That's what created multi-kilobyte memcpy() implementations, which barely beat REP MOVSB but cause huge icache bloat.

Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact