Torbjörn Granlund tg at gmplib.org
Thu Sep 5 09:27:59 UTC 2019

nisse at lysator.liu.se (Niels Möller) writes:

  For tuned *_METHOD values, it might be useful to let tuneup also add a
  comment with next best method and margin, something like

  #define HGCD2_METHOD 2 /* 3.2% faster than method 1 */

That'd be good!

  I did one variant with table lookup and an approximative quotient, which
  implies that one needed conditional swaps depending on a < b. That
  didn't beat plain division in my testing. But I also expect that we can
  beat plain division on more machines.

My suggested code was not 6 times faster!  I was fooled by inlining.  It
is 2-3 times faster only (on the system shell).

Please encrypt, key id 0xC8601622

More information about the gmp-devel mailing list