Hi Alex, Baseline: average benchmark over ten runs enc: 7948, dec: 5368 (blk/s)
> Another curiosity: I wonder if some `inline` declarations would help > tighten up the compiled code. Most of the word definitions are pretty > short, and are involved in a lot of loops. inline GF(2^8) arithmetic enc: 9071, dec: 7460 (blk/s) > The only things I notice now are just curiosities. I can see a lot of > `map`s and such involved, and wonder if you could gain any performance by > using the destructive equivalents. destructive operations enc: 10363, dec: 8945 (blk/s) Note that I reversed the order of your optimisations, because the improvement of destructive maps only kicks-in when inlining is performed. And also, destructive operations are dangerous: I got them wrong on first try! > Speaking of which, I don't think `inv-sub-bytes` gets typed like `sub-bytes` > (an untyped `[ inv-sbox nth ] map` vs the typed `sub-word`). In case it makes > a difference. :-) It doesn't, but I changed it for consistency (it doesn't hurt either). Thanks again, -- Gabriel ------------------------------------------------------------------------------ How ServiceNow helps IT people transform IT departments: 1. A cloud service to automate IT design, transition and operations 2. Dashboards that offer high-level views of enterprise services 3. A single system of record for all IT processes http://p.sf.net/sfu/servicenow-d2d-j _______________________________________________ Factor-talk mailing list Factor-talk@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/factor-talk
