infinite loop in mpz_urandomm with small third argument

[Torbjorn Granlund]

Camm Maguire <camm at enhanced.com> writes:

  Greetings!  We are considering using gmp random numbers in GCL.  I've
  run across an infinite loop with a random base of, e.g. 18.  Is this a
  known issue (current Debian unstable gmp-4.1.4)?  Right now we're
  working around by putting a floor under the base of (1<<31)-1.  I can
  provide a more detailed report if useful, but am out of time right
  now.
  
If you think you have found a bug in GMP, you should follow
the instructions in the manual on how to report it.

It is always a good idea to assume bugs are unknown.

--
Torbjörn