2 ms·
It still is in general situations (i.e. not the microbenchmarks where the ridiculously bloated unrolled "optimised" implementations may have a very slight edge.
by userbinator 4y ago
It still is in general situations (i.e. not the microbenchmarks where the ridiculously bloated unrolled "optimised" implementations may have a very slight edge.) I believe the Linux kernel uses it for this reason.
- jabl 4y agoThe kernel is a bit of a special case since very likely a syscall starts off with a cold I$, and also there's a lot of extra overhead if you insist on using SIMD registers. In general I agree with you though, optimizing memcpy implementations only against microbenchmarks is dumb.
- stncls 4y agoAlso, using SIMD registers is (generally) forbidden in kernel code, which heavily narrows down the competitors to "rep stosb".