The new dc_divappr_q_n ran for 16 hours or something like that,
without any problems. So it seems to be ok implementation-wise.

Bill.

2009/12/6 Bill Hart <[email protected]>:
> I've committed a "new" version of dc_divappr_q_n. I've done the following:
>
> 1) Changed the algorithm slightly (not that different to what was
> already there, but basically what I described to Brian and Jason about
> a month ago).
>
> 2) Added some comments giving an indication of why the algorithm works.
>
> 3) Figured out how to trigger the case at the end of the algorithm
> which we couldn't trigger with the test suite previously and checked
> that it works in that case.
>
> 4) Added a note that the current incarnation is not the original
> author's responsibility.
>
> 5) Begun extensive testing on Selmer using lt_try and the mpn test suite.
>
> We will need to run a complete new set of tests of MPIR on our test
> machines now that this has all been done, but let's wait until the
> next release candidate is done before starting that. We still have the
> next_prime thing to sort out and a few other bits and pieces.
>
> Bill
>

--

You received this message because you are subscribed to the Google Groups 
"mpir-devel" group.
To post to this group, send email to [email protected].
To unsubscribe from this group, send email to 
[email protected].
For more options, visit this group at 
http://groups.google.com/group/mpir-devel?hl=en.


Reply via email to