Re:tolower() in bulk at speed
by adrian_b on Tuesday June 28, 2022 @03:56AM
(#31905275)
Actually most modern SIMD instructions have a throughput of more than one per clock cycle.
Simple SIMD instructions, like integer addition, comparison, XOR, have a throughput of 2 per cycle (many ARM cores), 3 per cycle (most Intel CPUs) or even 4 per cycle (AMD Zen 3).
Similar simple 64-bit instructions have a higher throughput in instructions per cycle, from 3 per cycle up to 5 per cycle (Alder Lake), but when considering the register size the total throughput of the SIMD instructions is always higher.
Simple SIMD instructions, like integer addition, comparison, XOR, have a throughput of 2 per cycle (many ARM cores), 3 per cycle (most Intel CPUs) or even 4 per cycle (AMD Zen 3).
Similar simple 64-bit instructions have a higher throughput in instructions per cycle, from 3 per cycle up to 5 per cycle (Alder Lake), but when considering the register size the total throughput of the SIMD instructions is always higher.