On 19/12/17 08:54, Ingo Molnar wrote: > > * Eric Biggers <ebiggers3@xxxxxxxxx> wrote: > >> There may be a small overhead caused by replacing 'xchg REG, REG' with >> the needed sequence 'mov MEM, REG; mov REG, MEM; mov REG, REG' once per >> round. But, counterintuitively, when I tested "ctr-twofish-3way" on a >> Haswell processor, the new version was actually about 2% faster. >> (Perhaps 'xchg' is not as well optimized as plain moves.) > > XCHG has implicit LOCK semantics on all x86 CPUs, so that's not a surprising > result I think. Exchanging 2 registers can be done without memory access via: xor reg1, reg2 xor reg2, reg1 xor reg1, reg2 Juergen