Tue Aug 2 07:34:00 GMT 2011
I have released a new version of the gmp library for arbitrary precision
arithmetic. The new release is the final release in the 4.3 series.
Upstream fixes for issues in 4.3.1 http://gmplib.org/oldrel/
* The function mpf_eq has several bugs.
* For extremely large operands, there are overflow issues in
input conversion affecting mpz_set_str, mpz_inp_str,
mpf_set_str, and mpf_get_str.
* Multiplication of an operand of up to a few thousand digits
by a huge operand triggers a too large stack allocation,
leading to segfaults.
* A hard-to-trigger bug in FFT multiplication exists.
More information about the Cygwin-announce