[Botan-devel] Possible optimization if anyone wants to try

Jack Lloyd lloyd at randombit.net
Mon Jun 4 09:14:28 EDT 2007

Just a thought that occured to me - it's possible some low-level
BigInt loops could be reordered to decrease cache misses. I'm not
worried about page faults much, since the largest keys we might
practically use (8K?) is still much smaller than a single page.
However cache lines are usually 64 bytes, so it's possible there is
some cache misses that can be avoided.

This is mostly a note for myself to remember to look into this at some
point, but if someone is interested feel free to take a look.


More information about the botan-devel mailing list