Re:XXH3 – a new speed-optimized hash algorithm
by andromeduck on Friday March 15, 2019 @06:05PM
(#19404752)
Addition and multiplication also leak more information via side channels than bit shift and xor.
[ Parent
| on HN ]
Re:XXH3 – a new speed-optimized hash algorithm
by dragontamer on Friday March 15, 2019 @06:17PM
(#19404838)
I'd be surprised if that's true.
I'm pretty sure addition is always 1-clock tick latency / 3-instructions per cycle. While Multiplication is 5-clock tick latency / 1-instruction per cycle IIRC.
So if the timing of those instructions are constant, it seems difficult to side-channel / timing attack them.
Re:XXH3 – a new speed-optimized hash algorithm
by andromeduck on Monday March 18, 2019 @07:57PM
(#19426612)
there's also power, 0xffffffff + 0x1 results in 32 bit flips, 0x0 + 0x1 results in 1, they don't consume the same amount of energy.