mpn_mul_fft type overflow issue

[Mark Sofroniou]

I've cleaned this up some more. I'll send you the new patch in
a few days once I have fully tested it.

I'll also add an mpn_tdiv_qr inverse test to check large multiplications 
- it
should at least be a basic check for type overflow in that code.

On 09/18/2013 01:15 PM, Torbjorn Granlund wrote:
>    There are things like:
>    
>      int sqr = (n == m && nl == ml);
>    
>    that would maybe be clearer as a boolean type.
>    
> Let's leave that for a different patch.  (You may notice that the
> conditions for sqr vary with file, something that we might want to fix
> too.)
>
Agreed. Changes like that should probably go into a development
version so they don't interfere with any workflow in the 5.1.2 branch.

Mark