Re:tolower() in bulk at speed
by moonchild on Monday June 27, 2022 @09:20PM
(#31903018)
Only port 5 and port 0/1 can do avx512 instructions on skylake/icelake, so I don't think you can get better throughput than 0.5 on current parts. Unless you count load & store as well, like I mentioned.
[ Parent
| on HN ]
(Unless you count an avx512-specific instruction operating on a 32- or 16-byte vector as an 'avx512 instruction'.)
Skylake is 7 years old
And icelake is 3 years old. If you have a newer avx512-supported chip which can do more than two 512-bit alu ops per cycle, I would love to take a look at it.