#11666: Upgrade MPFR to 3.1.0
--------------------------------+-------------------------------------------
Reporter: leif | Owner: leif
Type: enhancement | Status: needs_info
Priority: major | Milestone: sage-5.0
Component: packages | Keywords: sd32 MPFR spkg wishlist
sd35.5
Work_issues: correct doctests | Upstream: N/A
Reviewer: Paul Zimmermann | Author: Mike Hansen, Jean-Pierre
Flori
Merged: | Dependencies: #12171
--------------------------------+-------------------------------------------
Comment(by jpflori):
I've had a look at the random MPFR real number generation code and there
was indeed a "hack" for 32 bits systems before which requested more bits
(31 from randstate doc, although real_mpfr says 32) from MPIR to get an
equivalent MPFR behavior. Although, if I force sage to do the same on my
64 bits system (just adding a call to rstate.c_random() unconditionally in
real_mpfr), I do not get what 32 bits (and 64 bits) systems used to return
(nor what my 64 bits system now return which was to expect).
Paul, I'm confused by your last message. By vanilla 4.8, do you mean
vanilla+this ticket and dependencies? If that's the case, and I somehow
understand correctly the random generation stuff of MPIR and MPFR which
should both be hardware independent now, the change I made above should
give me a similar behavior as you on a 32 bits system and get no doctest
failures.
For example the last failing test in real_mpfr used to give
{{{
0.979095507956490
}}}
With this ticket and dependencies, I get (on all of my 64 systems)
{{{
-0.616678906367394
}}}
With the additional call to c_random() I get (on all of my 64 systems)
{{{
0.681934947736557
}}}
--
Ticket URL: <http://trac.sagemath.org/sage_trac/ticket/11666#comment:45>
Sage <http://www.sagemath.org>
Sage: Creating a Viable Open Source Alternative to Magma, Maple, Mathematica,
and MATLAB
--
You received this message because you are subscribed to the Google Groups
"sage-trac" 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/sage-trac?hl=en.
