0 0 0x01 -1 0 3097903805 1471435219981
implementation   native         byteswap       
scalar           8913241388     8093205981     
superscalar      13335571664    11851318271    
superscalar4     13836068980    11271172898    
sse2             21073198181    11638989917    
ssse3            21056976228    19743661680    
avx2             41577944073    34889828558    
fastest          avx2           avx2           
