> this is a typical symptom of trying to link a wrong version of libpng.
>
Yes, but why does the `sage -f libpng` was not forced since it was
necessary? I am not an expert at all on this, but from my point of view, it
is like if there was problem with dependencies of giac.
>
> But what
On Thu, Mar 15, 2018 at 7:26 AM, Jeroen Demeyer wrote:
> On 2018-03-14 19:02, Erik Bray wrote:
>>
>> two Integers' mpz_t structs' _mp_d
>> member (the array of limbs) is being set to the same memory address.
>
>
> You really mean two different Integer objects with the same
On Wed, Mar 14, 2018 at 10:03 PM, Vincent Delecroix
<20100.delecr...@gmail.com> wrote:
> Congratulations Erik! You found the field with one element in SageMath:
>
> https://en.wikipedia.org/wiki/Field_with_one_element
>
> I am really impressed.
Hah! Yes, I was extremely confused as to why Sage
On Wed, Mar 14, 2018 at 8:33 PM, Jeroen Demeyer wrote:
> On 2018-03-14 19:02, Erik Bray wrote:
>>
>> I see now--less likely a bug with MPIR/GMP itself, and more likely a
>> bug with Sage's fast_tp_new stuff for Integer.
>
>
> I don't know whether it will fix that bug or not,
On 2018-03-14 19:02, Erik Bray wrote:
two Integers' mpz_t structs' _mp_d
member (the array of limbs) is being set to the same memory address.
You really mean two different Integer objects with the same _mp_d (as
opposed to the same Integer object being used mistakenly in two
different