Torbjörn Granlund tg at gmplib.org
Sat Jul 3 08:38:59 UTC 2021

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

  Looks better, but few tuneup results yet. Method 3 wins on a
  few machines (https://gmplib.org/devel/thres/DIV_QR_1N_PI1_METHOD), with
  all of method 1, 2 and 4 appearing as runner up on some machine.

The presentation in those "/thres/" pages is a bit strange when it is
not operand size cutoff values but, like here, methods.  I.e., measured
method 3 vs configured method is marked as a major difference.  And the
average is 0.975...

I suppose it would be somewhat interesting to know how close the
measured best and measured 2nd best methods are, or how much better the
measured best methods is than the configured method.

And, when there is asm, the scary performance ratio between C and asm
which would suggest that C-to-C comparisons of tight GMP loops might not
be terribly relevant.  :-)

Please encrypt, key id 0xC8601622

More information about the gmp-devel mailing list