From: terje.mathisen@nospicedham.tmsw.no   
      
   James Van Buskirk wrote:   
   > Just for fun I thought I would try various strategies for permuting   
   > an array of single-precision floating point numbers using AVX2.   
   > bitrev1.asm just does vpunpckl/hdq/qdq or its lane-crossing   
   > synthesis with vperm2i128 and has a hard limit of 24 clocks to   
   > bit-reverse a 64-element array because all of these operations   
   > use pipeline 5.   
   [snip]   
   > To make a long story short these tests seemed to show that   
   > bitrev1 took about 27 clocks, bitrev2 about 23, bitrev3 about   
   > 93, and bitrev4 about 72 clocks.   
      
   Interesting stuff, thanks for posting!   
      
   >   
   > Oh well, maybe I should have gone out and played in the snow   
   > instead today.   
      
   Snow today, in late May? Are you in some New Zealand southern island   
   mountains/ditto South America/Antarctica?   
      
   Or just high up in the Rockies (US/Canada) and the snow is old stuff?   
      
   Terje   
      
   --   
   -    
   "almost all programming can be viewed as an exercise in caching"   
      
   --- SoupGate-Win32 v1.05   
    * Origin: you cannot sedate... all the things you hate (1:229/2)