I think the following are problems:

-----------------

In routine MParith::mpmov :

wrong: for(j=n_min;j<n-1;j++) u[j]=0;
right: for(j=n_min;j<n;j++) u[j]=0;

-----------------

In description of routine MParith::mpsqrt, it says:

"u[0..n-1] is set to the reciprocal [of the sqrt] (radix point BEFORE u[0])."

However, it should say: "(radix point AFTER u[0])"

-----------------

In description of routine MParith::mp2dfr, it says:

"Converts a radix 256 fraction a[0..n-1] (radix point BEFORE a[0]) to a decimal fraction represented as an ASCII string"

However, it should say: "(radix point AFTER a[0])"

-----------------