Martin R. Albrecht

Results 159 comments of Martin R. Albrecht

Something is quite off with our heuristics: ``` sage: from estimator import * sage: for k in (20, 30, 40, 50, 60, 70, 80, 90): ....: print(k, repr(LWE.coded_bkw(LWE.Parameters(1024, 2**k, ND.Uniform(-1,...

I pushed something for this particular crash. Still, I wouldn't trust the output: ```python sage: from estimator import * sage: for k in (20, 30, 40, 50, 60, 70, 80,...

Mhhh, hard to debug from afar. I'm still on GCC 12.x though.

I could cut a new release if it helps?

Okay, we got CI set up now. Can you trigger it by making some push to your PR branch?

There seems to be a conflict between Cython < 3 and Cython 3 on how this function should be declared: https://github.com/fplll/fpylll/blob/21123b5a88a5d0672f172b824925b00e3839b7c0/src/fpylll/fplll/enumeration_callback_helper.h#L8C1-L11C5 I have no good idea how to handle this?...

Is `gmpy2` *the* `mpz_t` wrapper to use? I'm a bit hesitant to add more dependencies.

I wonder if we could add Marc's idea though: an interface that takes an input and repeatedly considers it mod MAX_INT_SIZE to construct the internal integer? This then may or...

It seems cysignals fails with a current Python (3.11) and Cython (3.0). In addition FPyLLL Failed, but that might be fixed now.

See https://github.com/sagemath/cysignals/pull/174