I've just implemented ARM NEON support for JtR's bitslice DES. Only ~35% faster than non-NEON on Cortex-A9 in Zynq: http://www.openwall.com/lists/john-users/2013/06/03/1 …
@solardiz That sounds pretty impressive, what speedup were you expecting?
-
-
@taviso Before reading the docs on Cortex-A9's instruction timings (turn out to be very poor for NEON), I naively expected a ~4x speedupThanks. Twitter will use this to make your timeline better. UndoUndo
-
-
-
@taviso I think a much higher (than this ~35%) speedup is possible for MD*/SHA* by interleaving 2+ sets of instances to hide latenciesThanks. Twitter will use this to make your timeline better. UndoUndo
-
Loading seems to be taking a while.
Twitter may be over capacity or experiencing a momentary hiccup. Try again or visit Twitter Status for more information.