Re:KolibriOS – operating system written entirely in assembly...
by pjmlp on Wednesday April 15, 2020 @09:35AM
(#22877495)
There are some LLVM optimization talks that show otherwise, where longer vectorized code actually runs faster than the smaller version, though.
[ Parent
| on HN ]
Re:KolibriOS – operating system written entirely in assembly...
by userbinator on Wednesday April 15, 2020 @09:46AM
(#22877622)
...in microbenchmarks.
That's what created multi-kilobyte memcpy() implementations, which barely beat REP MOVSB but cause huge icache bloat.