Quote:
Originally Posted by Jean PennĂ©
Hi,
These two bugs are now fixed :
 For 10^500+961, it is really a bug of the strong Fermat PRP code in the Llr.c file. It will be fixed in the Development version ; here is the correct behaviour :
Starting probable prime test of 10^500+961
Using allcomplex FFT length 192, a = 3
10^500+961 is base 3Strong Fermat PRP! Time : 121.052 ms.
Starting Lucas sequence
Using allcomplex FFT length 192, P = 5, Q = 3
10^500+961 is strongFermat and Lucas PRP, Starting Frobenius test sequence
Using allcomplex FFT length 192, Q = 3
10^500+961 is strongFermat, Lucas and Frobenius PRP! (P = 5, Q = 3, D = 13) Time : 72.780 ms.
For 5*10^19737+3, it is a gwnum problem : gwnum chooses a fullcomplex 4K FFT, which seems to be too small, and causes a segmentation fault after a few iterations...

I suspect that it might be CPU specific or specific to the 32bit version. On 64bit MacIntel, pfgw chooses the 4K FFT and it runs without a problem. Granted pfgw and llr use different code. I would have to try with 64bit llr on MacIntel to know for certain.