Currently ASSERT_ZERO(n) is not taking n % p and thus would return false e.g. for n = p (which is in fact 0 % p).
This change converts the value we compare to 0 into an mpz_t (which does the modulo operation) and adds a slightly more convenient error message, in case the assert fails.
Currently ASSERT_ZERO(n) is not taking n % p and thus would return false e.g. for n = p (which is in fact 0 % p).
This change converts the value we compare to 0 into an mpz_t (which does the modulo operation) and adds a slightly more convenient error message, in case the assert fails.