[PATCH] ipsum_calc_block: Optimize size and speed

Martin Mares mj at ucw.cz
Fri Apr 23 19:23:18 CEST 2010


> > > So there isn't really difference in performance of both
> > > implementations. Even on slow embedded AMD Geode CPU, it gives
> > > ~ 180 MB/s.
> No difference? what does 1.2 mean? to me this means 20% which is a lot

Yes, but according to Santiago's benchmarks, your code is sometimes 20%
faster, sometimes 20% slower. It does not seem like a reason for change.

If you have any benchmark showing that BIRD spends a substantial amount
of time in this function, let's optimize it properly, even if it means
having several versions for different CPU's. Otherwise, let us stick
with the current code and concentrate our effort on places which matter.

The difference in code size and in readability is really tiny.

				Have a nice fortnight
Martin `MJ' Mares                          <mj at ucw.cz>   http://mj.ucw.cz/
Faculty of Math and Physics, Charles University, Prague, Czech Rep., Earth
Why is "abbreviation" such a long word?

More information about the Bird-users mailing list