Bill Allombert on Fri, 04 Jun 2004 12:43:42 +0200


[Date Prev] [Date Next] [Thread Prev] [Thread Next] [Date Index] [Thread Index]

Re: round4 performance


On Thu, Jun 03, 2004 at 07:21:45PM +0200, Karim Belabas wrote:
> I have corrected the SEGV above and went on fixing the above-mentioned 2
> inefficiencies.  The code is almost entirely modular now, and often quite a
> bit faster (~ a factor 2 for the big polynomial). Esp. in the large
> degrees you seem to favor :-).
> 
> The current code passes my test-suite [ make test-round4 + all polynomials
> submitted so far in this thread ].
> 
> Please re-check ! I'll try and cleanup everything after that last checkpoint.

That pass my test suite in 142s with 4M stack instead of
249s and 8Mb. Good job!

Cheers,
Bill.