SIGABRT when taking the multiplicative inverse of the additive identity in a finite field

[Demi Obenour]

Consider the following Sage code

#!/usr/bin/env sage

FiniteField(2^128)()^-1

This crashes Sage with SIGABRT.  While there is no answer mathematically, this should throw an exception, not cause a C-level crash.  (Alternatively, one could return 0 here; this is what the Rijndael S-box does.)

[Dima Pasechnik]

yes, it's a bug, I've opened https://trac.sagemath.org/ticket/23276

to deal with it.