Mercurial > hg > xemacs-beta

comparison man/xemacs-faq.texi @ 5739:a2912073be85

Find changesets by keywords (author, files, the commit message), revision number or hash, or revset expression.
Support bignums with MPIR. Add documentation on the bignum, ratio, and bigfloat implementations. See xemacs-patches message with ID <CAHCOHQkytZao7Uk9ggeo1HKKJtN1bqO054X2mPsGYyQFjbHrZA@mail.gmail.com> and following messages.
author Jerry James <james@xemacs.org>
date Wed, 19 Jun 2013 09:30:30 -0600
parents 40a52efbf3a3
children d11efddf3617
comparison
equal deleted inserted replaced
5738:f6af091ac654 5739:a2912073be85
7948 @node Q7.2.1, Q7.2.2, Q7.1.11, Advanced 7948 @node Q7.2.1, Q7.2.2, Q7.1.11, Advanced
7949 @unnumberedsubsec Q7.2.1: What are bignums, ratios, and bigfloats in Lisp? 7949 @unnumberedsubsec Q7.2.1: What are bignums, ratios, and bigfloats in Lisp?
7950 7950
7951 Thanks to @email{james@@xemacs.org, Jerry James}, XEmacs 21.5.18 and 7951 Thanks to @email{james@@xemacs.org, Jerry James}, XEmacs 21.5.18 and
7952 later can use the capabilities of multiple-precision libraries that may 7952 later can use the capabilities of multiple-precision libraries that may
7953 be available for your platform. The GNU Multiple Precision (GMP) and 7953 be available for your platform. The GNU Multiple Precision (GMP),
7954 BSD Multiple Precision (MP) libraries are partially supported. GMP 7954 Multiple Precision Integers and Rationals (MPIR), and BSD Multiple
7955 gives you @dfn{bignums} (arbitrary precision integers), @dfn{ratios} 7955 Precision (MP) libraries are supported. GMP and MPIR give you
7956 (arbitrary precision fractions), and @dfn{bigfloats} (arbitrary 7956 @dfn{bignums} (arbitrary precision integers), @dfn{ratios} (arbitrary
7957 precision floating point numbers). GNU MP is better-supported by XEmacs 7957 precision fractions), and @dfn{bigfloats} (arbitrary precision floating
7958 at the time of writing (2004-04-06). BSD MP support does not include 7958 point numbers). GMP and MPIR are better-supported by XEmacs. BSD MP
7959 ratios or bigfloats, and it throws errors that aren't understood. 7959 support does not include ratios or bigfloats.
7960 7960
7961 In most cases, bignum support should be transparent to users and Lisp 7961 In most cases, bignum support should be transparent to users and Lisp
7962 programmers. A bignum-enabled XEmacs will automatically convert from 7962 programmers. A bignum-enabled XEmacs will automatically convert from
7963 fixnums to bignums and back in pure integer arithmetic, and for GNU MP, 7963 fixnums to bignums and back in pure integer arithmetic, and for GMP and
7964 from floats to bigfloats. (Bigfloats must be explicitly coerced to 7964 MPIR, from floats to bigfloats. (Bigfloats must be explicitly coerced
7965 other types, even if they are exactly representable by less precise 7965 to other types, even if they are exactly representable by less precise
7966 types.) The Lisp reader and printer have been enhanced to handle 7966 types.) The Lisp reader and printer have been enhanced to handle
7967 bignums, as have the mathematical functions. Rationals (fixnums, 7967 bignums, as have the mathematical functions. Rationals (fixnums,
7968 bignums, and ratios) are printed using the @samp{%d}, @samp{%o}, 7968 bignums, and ratios) are printed using the @samp{%d}, @samp{%o},
7969 @samp{%x}, and @samp{%u} format conversions. The read syntax for ratios 7969 @samp{%x}, and @samp{%u} format conversions. The read syntax for ratios
7970 is @samp{3/5}. 7970 is @samp{3/5}.
8000 can't be ruled out. ``Arbitrary'' precision means precisely what it 8000 can't be ruled out. ``Arbitrary'' precision means precisely what it
8001 says. If you work with extremely large numbers, your machine may 8001 says. If you work with extremely large numbers, your machine may
8002 arbitrarily decide to hand you an unpleasant surprise rather than a 8002 arbitrarily decide to hand you an unpleasant surprise rather than a
8003 bignum @ref{Q7.2.2, XEmacs segfaults when I use very big numbers!}. 8003 bignum @ref{Q7.2.2, XEmacs segfaults when I use very big numbers!}.
8004 8004
8005 To configure with GNU MP, add @samp{--use-number-lib=gmp} 8005 To configure with GMP, add @samp{--enable-bignum=gmp} to your invocation
8006 (@samp{--enable-bignum=gmp} in 21.5 or later) to your invocation of 8006 of @file{configure}. For MPIR, use @samp{--enable-bignum=mpir}. For
8007 @file{configure}. For BSD MP, use @samp{--use-number-lib=mp} 8007 BSD MP, use @samp{--enable-bignum=mp}.
8008 (@samp{--enable-bignum=mp} for 21.5).
8009
8010 If you would like to help with bignum support, especially on BSD MP,
8011 please subscribe to the @uref{http://www.xemacs.org/Lists/#xemacs-beta,
8012 XEmacs Beta mailing list}, and book up on @file{number-gmp.h} and
8013 @file{number-mp.h}. Jerry has promised to write internals documentation
8014 eventually, but if your skills run more to analysis and documentation
8015 than to writing new code, feel free to fill in the gap!
8016 8008
8017 8009
8018 @node Q7.2.2, Q7.2.3, Q7.2.1, Advanced 8010 @node Q7.2.2, Q7.2.3, Q7.2.1, Advanced
8019 @unnumberedsubsec Q7.2.2: XEmacs segfaults when I use very big numbers! 8011 @unnumberedsubsec Q7.2.2: XEmacs segfaults when I use very big numbers!
8020 8012
8034 8026
8035 8027
8036 @node Q7.2.3, Q7.2.4, Q7.2.2, Advanced 8028 @node Q7.2.3, Q7.2.4, Q7.2.2, Advanced
8037 @unnumberedsubsec Q7.2.3: Bignums are really slow! 8029 @unnumberedsubsec Q7.2.3: Bignums are really slow!
8038 8030
8039 Many Linux distributions compile all their packages for the i386, and 8031 Many Unix and Linux distributions compile all packages for a generic
8040 this is costly. An optimized version can give you two or three orders 8032 version of the supported CPU, and this is costly. An optimized version
8041 of magnitude better performance for a Pentium III or IV. (Yes, really. 8033 can improve responiveness dramatically; see
8042 See @uref{http://www.swox.com/gmp/gmp-speed.html}.) 8034 @uref{http://gmplib.org/gmpbench.html}.)
8043 8035
8044 8036
8045 @node Q7.2.4, , Q7.2.3, Advanced 8037 @node Q7.2.4, , Q7.2.3, Advanced
8046 @unnumberedsubsec Q7.2.4: Equal bignums don't compare as equal! What gives? 8038 @unnumberedsubsec Q7.2.4: Equal bignums don't compare as equal! What gives?
8047 8039