individual logs fail for full_p30_JL
The output of the full_p30_JL
test says:
Info:root: Checking that log(2) and log(3) are consistent...
Info:root: p = 191907783019725260605646959711
Info:root: ell = 101538509534246169632617439
Info:root: log2 = 0
Info:root: log3 = 0
Info:root: If you want to compute a new target, run ./cado-nfs.py /tmp/cado.cudjbmpx/p30.parameters_snapshot.0 target=<target>
To proceed, we need the descent_init_Fp
binary, in fact. In turn, this requires ecm, and this is something that we should say.
Assuming that we succeed in building descent_init_Fp
, actually doing a the message suggests takes an awful lot of time (close to a minute), and then the script crashes.
Info:Complete Factorization / Discrete logarithm: Total cpu/elapsed time for entire discrete log: 333.18/112.607
Info:root: Checking that log(2) and log(3) are consistent...
Info:root: p = 191907783019725260605646959711
Info:root: ell = 101538509534246169632617439
Info:root: log2 = 0
Info:root: log3 = 0
Info:root: Also check log(target) vs log(2) ...
Traceback (most recent call last):
File "/home/thome/NFS/cado/./cado-nfs.py", line 154, in <module>
assert pow(int(target), log2*((p-1) // ell), p) == pow(2, logtarget*((p-1) // ell), p)
AssertionError
Both the message flakiness, and the crash, should be fixed.