diff options
| author | vlefevre <vlefevre@280ebfd0-de03-0410-8827-d642c229c3f4> | 2007-01-10 15:40:05 +0000 |
|---|---|---|
| committer | vlefevre <vlefevre@280ebfd0-de03-0410-8827-d642c229c3f4> | 2007-01-10 15:40:05 +0000 |
| commit | 7b17d6f4be472cf6b78ca8172c911d5a7646b174 (patch) | |
| tree | 743f7ce16c381d580bb7af8780b5bb1209329c17 /FAQ.html | |
| parent | 5ccc1c8b25da336f375811eeb0731b5ba1df1975 (diff) | |
| download | mpfr-7b17d6f4be472cf6b78ca8172c911d5a7646b174.tar.gz | |
FAQ.html: Added details to MPF-related Q&A's.
git-svn-id: svn://scm.gforge.inria.fr/svn/mpfr/trunk@4327 280ebfd0-de03-0410-8827-d642c229c3f4
Diffstat (limited to 'FAQ.html')
| -rw-r--r-- | FAQ.html | 16 |
1 files changed, 13 insertions, 3 deletions
@@ -2,7 +2,7 @@ <!DOCTYPE html PUBLIC "-//W3C//DTD XHTML 1.0 Strict//EN" "http://www.w3.org/TR/xhtml1/DTD/xhtml1-strict.dtd"> <!-- -Copyright 2000, 2001, 2002, 2003, 2004, 2005, 2006 Free Software Foundation, Inc. +Copyright 2000, 2001, 2002, 2003, 2004, 2005, 2006, 2007 Free Software Foundation, Inc. Contributed by the Spaces project, INRIA Lorraine. This file is part of the MPFR Library. @@ -47,7 +47,6 @@ dt margin-top: 2ex; margin-bottom: 1ex; font-weight: bolder; - background: #dddddd; } /* For testing: dd { background: #ddddff } */ @@ -82,6 +81,8 @@ li > p, dd > p dl.faq { counter-reset: faq } +dl.faq dt { background: #dddddd } + dl.faq dt:before { counter-increment: faq; @@ -117,6 +118,13 @@ is a minimum value (<cite><acronym>MPF</acronym></cite> generally uses a higher precision). With the additional difference below, this implies that the <cite><acronym>MPFR</acronym></cite> results do not depend on the number of bits (16, 32, 64 or more) of the underlying architecture.</p></li> +<li><p>As a consequence, <cite><acronym>MPFR</acronym></cite> uses a +base-2 exponent, whereas in <cite><acronym>MPF</acronym></cite>, this +is a base-2<sup>32</sup> or base-2<sup>64</sup> exponent, depending on +the limb size. For this reason (and other internal ones), the maximum +exponent range in <cite><acronym>MPFR</acronym></cite> is different +(and smaller, if the exponent is represented by the same type as in +<cite><acronym>MPF</acronym></cite>).</p></li> <li><p><cite><acronym>MPFR</acronym></cite> provides an additional rounding mode argument to its functions; furthermore, it is guaranteed that the result of any operation is the nearest possible floating-point value from @@ -148,7 +156,9 @@ mode. Thus you simply need to add the following line in all your files using <cite><acronym>MPF</acronym></cite> functions: <code class="block-code">#include <mpf2mpfr.h></code> just after the <samp>gmp.h</samp> and <samp>mpfr.h</samp> -header files.</p></dd> +header files. If the program uses <cite><acronym>MPF</acronym></cite> +internals (such as direct access to <code>__mpf_struct</code> members), +additional changes will be needed.</p></dd> <dt id="undef_ref">When I link my program with <cite><acronym>MPFR</acronym></cite>, I get undefined reference |