Cofactor canonicalisation of mpn_gcdext

[Torbjorn Granlund]

Marc Glisse <marc.glisse at inria.fr> writes:

  On Wed, 4 May 2011, Niels Möller wrote:
  
  > nisse at lysator.liu.se (Niels Möller) writes:
  >
  >> mpz_gcdext still allocates un limbs for both gp and sp (and did allocate
  >> un+1 until my checkin yesterday). I'll try reducing that to vsize, to
  >> see what happens (and gsize == usize actually happens in the case vsize
  >> == 0, which is allowed by mpz_gcdext but not by mpn_gcdext).
  >
  > I'm about to check in the attached mpz/gcdext.c.
  
  In 32 bits, the mpz_gcdext test currently fails with:
  
  ERROR in test 136
  mpz_gcdext returned invalid result
  op1=0
  op2=0
  mpz_gcdext returns:
  0
  s=1
  /bin/sh: line 5:  3298 Aborted                 ${dir}$tst
  FAIL: t-gcd

I cannot reproduce this on shell (using ABI=32).

-- 
Torbjörn