Array Reverse with SSE

[frktons]

I think I reached the bottom line of SSSE3 code:

------------------------------------------------------------------------
Intel(R) Core(TM)2 CPU          6600  @ 2.40GHz

Instructions: MMX, SSE1, SSE2, SSE3, SSSE3
------------------------------------------------------------------------
1.191  cycles for Reverse Array with PSHUFB
2.064  cycles for Reverse Array with MOV/BSWAP
1.617  cycles for Reverse Array with MOV/BSWAP with 4 GPRs
   565  cycles for Reverse Array with PSHUFB using 4 xmm unrolled 4
   594  cycles for Reverse Array with PSHUFB using 4 xmm
------------------------------------------------------------------------
   794  cycles for Reverse Array with PSHUFB
2.064  cycles for Reverse Array with MOV/BSWAP
1.639  cycles for Reverse Array with MOV/BSWAP with 4 GPRs
   565  cycles for Reverse Array with PSHUFB using 4 xmm unrolled 4
   596  cycles for Reverse Array with PSHUFB using 4 xmm
------------------------------------------------------------------------

565 CPU Cycles seems to be the lowest reachable value on my system.
While other routines change performance show every time, using 4 xmm
unrolled 4 times tends to give always the same value.