| Here is a summary of the entry points in the MP package.
An asterisk denotes routines that are most useful to the caller.
* EXAMPLE A SMALL MAIN PROGRAM GIVING AN EXAMPLE OF THE USE OF MP.
* MPABS COMPUTES ABSOLUTE VALUE OF AN MP NUMBER
CALL MPABS (X, Y) MEANS Y = ABS(X)
* MPADD ADDS TWO MP NUMBERS
CALL MPADD (X, Y, Z) MEANS Z = X + Y
* MPADDI ADDS AN MP NUMBER TO AN INTEGER,
GIVING A MULTIPLE-PRECISION RESULT, SPACE = 2T+6
CALL MPADDI (X, IY, Z) MEANS Z = X + IY
* MPADDQ ADDS A RATIONAL NUMBER TO AN MP NUMBER,
SPACE = 2T+6
CALL MPADDQ (X, I, J, Y) MEANS Y = X + I/J
MPADD2 ROUTINE CALLED BY MPADD AND MPSUB
MPADD3 ROUTINE CALLED BY MPADD2
MPART1 COMPUTES ARCTAN(1/N) FOR N .GT. 1 (CALLED BY MPPI)
SPACE = 2T+6
* MPASIN COMPUTES ARCSIN OF AN MP NUMBER,
USING AN O(M(T)T) METHOD, SPACE = 5T+12
CALL MPASIN (X, Y) MEANS Y = ARCSIN(X)
* MPATAN COMPUTES ARCTAN OF AN MP NUMBER
USING AN O(T.M(T)) ALGORITHM, SPACE = 5T+12
CALL MPATAN (X, Y) MEANS Y = ARCTAN(X)
* MPBERN COMPUTES BERNOULLI NUMBERS B2, B4, B6, ...
SPACE = 8T+18
* MPBESJ COMPUTES BESSEL FUNCTION J(NU,X) FOR MP X
AND SMALL INTEGER NU, SPACE = 14T+156
CALL MPBESJ (X, NU, Y) MEANS Y = J(NU,X)
MPBES2 ROUTINE CALLED BY MPBESJ (USES BACKWARD RECURRENCE
TO EVALUATE J(NU,X)), SPACE = 8T+18
MPCDM CONVERTS DOUBLE-PRECISION TO MULTIPLE-PRECISION
MPCHK PRINTS ERROR MESSAGE ON UNIT LUN IF B, T, M OR MXR
IS ILLEGAL, OR ON UNIT 6 IF LUN IS ILLEGAL
(LUN SHOULD BE IN RANGE 1 TO 99)
* MPCIM CONVERTS INTEGER TO MULTIPLE-PRECISION
CALL MPCIM (IX, Z) MEANS Z = IX
MPCLR SETS SOME DIGITS OF AN MP NUMBER TO ZERO
MPCMD CONVERTS AN MP NUMBER TO DOUBLE-PRECISION REAL
MPCMDE CONVERTS AN MP NUMBER TO (DOUBLE-PRECISION)
FRACTION AND (DECIMAL) EXPONENT,
SPACE = 6T+14
MPCMEF CONVERTS MP NUMBER TO FRACTION AND (DECIMAL)
EXPONENT, SPACE = 5T+12
* MPCMF FINDS FRACTIONAL PART OF AN MP NUMBER
MPCMI CONVERTS AN MP NUMBER TO AN INTEGER
* MPCMIM CONVERTS AN MP NUMBER TO A MULTIPLE-PRECISION INTEGER
* MPCMPA COMPARES ABSOLUTE VALUES OF TWO MP NUMBERS
MPCMPA (X, Y) RETURNS SIGN(ABS(X)-ABS(Y))
* MPCMPI COMPARES AN MP NUMBER WITH AN INTEGER, SPACE = 2T+6
MPCMPI (X, I) RETURNS SIGN(X-I)
* MPCMPR COMPARES AN MP NUMBER WITH A REAL, SPACE = 2T+6
MPCMPR (X, RI) RETURNS SIGN(X-RI)
* MPCMR CONVERTS AN MP NUMBER TO (SINGLE-PRECISION) REAL
CALL MPCMR (X, RZ) MEANS RZ = SNGL(X)
* MPCMRE CONVERTS AN MP NUMBER TO EXPONENT AND
(SINGLE-PRECISION) FRACTION, I.E. F*10**I
SPACE = 6T+14
* MPCOMP COMPARES TWO MP NUMBERS
MPCOMP (X, Y) RETURNS SIGN(X-Y)
* MPCOS COMPUTES COSINE OF AN MP NUMBER, USING AN
O(M(T)T/LOG(T)) METHOD, SPACE = 5T+12
CALL MPCOS (X, Y) MEANS Y = COS(X)
* MPCOSH COMPUTES HYPERBOLIC COSINE OF AN MP NUMBER
USING MPEXP, SPACE = 5T+12
CALL MPCOSH (X, Y) MEANS Y = COSH(X)
* MPCQM CONVERTS A RATIONAL NUMBER TO MULTIPLE-PRECISION
CALL MPCQM (I, J, Q) MEANS Q = I/J
MPCRM CONVERTS REAL TO MULTIPLE-PRECISION
CALL MPCRM (RX, Z) MEANS Z = RX
* MPDAW COMPUTES DAWSONS INTEGRAL, DAW(X) = EXP(-X**2)*(INTEGRAL
FROM 0 TO X OF EXP(U**2) DU), SPACE = 5T+17
CALL MPDAW (X, Y) MEANS Y = DAW(X)
* MPDIV DIVIDES TWO MP NUMBERS, SPACE = 4T+10
CALL MPDIV (X, Y, Z) MEANS Z = X/Y
* MPDIVI DIVIDES AN MP NUMBER BY AN INTEGER
USING AN O(T) METHOD (MUCH FASTER THAN MPDIV)
CALL MPDIVI (X, IY, Z) MEANS Z = X/IY
MPDUMP DUMPS AN MP NUMBER (USEFUL FOR DEBUGGING)
CALL MPDUMP (X) DUMPS THE MP NUMBER X ON UNIT LUN
* MPEI EVALUATES EXPONENTIAL INTEGRAL OF AN MP NUMBER,
SPACE = 19T+31
CALL MPEI (X, Y) MEANS Y = EI(X)
* MPEPS COMPUTES THE (MULTIPLE-PRECISION) MACHINE PRECISION
CALL MPEPS (X) MEANS X = 0.5*B**(1-T) IF B EVEN
* MPERF COMPUTES ERROR FUNCTION OF AN MP NUMBER,
SPACE = 5T+12
CALL MPERF (X, Y) MEANS Y = ERF(X)
* MPERFC COMPUTES COMPLEMENTARY ERROR FUNCTION OF AN MP NUMBER,
SPACE = 12T+26
CALL MPERFC (X, Y) MEANS Y = ERFC(X)
MPERF2 COMPUTES EXP(X*X)*(INTEGRAL FROM 0 TO X OF
EXP(-U*U) DU), CALLED BY MPERF, SPACE = 5T+12
MPERF3 ROUTINE CALLED BY MPERF, MPDAW AND MPERFC,
SPACE = 4T+10
MPERR ERROR HANDLING ROUTINE (TERMINATES EXECUTION AT
PRESENT BUT MAY EASILY BE MODIFIED).
* MPEUL RETURNS EULERS CONSTANT (GAMMA = 0.57721566...) TO
MULTIPLE-PRECISION ACCURACY, SPACE = 5T+14
CALL MPEUL (G) MEANS G = 0.57721566...
* MPEXP COMPUTES EXPONENTIAL OF A MULTIPLE-PRECISION
NUMBER, USING AN O(SQRT(T)M(T)) METHOD, SPACE = 4T+10
CALL MPEXP (X, Y) MEANS Y = EXP(X)
MPEXP1 COMPUTES EXP(X)-1 FOR ABS(X) .LT. 1 (CALLED BY
MPEXP, MPSINH AND MPTANH), SPACE = 3T+8
MPEXT A ROUNDING ROUTINE CALLED BY MPDIV AND MPSQRT
* MPGAM COMPUTES GAMMA FUNCTION OF AN MP ARGUMENT,
SPACE SAME AS FOR MPLNGM (IN WORST CASE)
CALL MPGAM (X, Y) MEANS Y = GAMMA(X)
MPGAMQ COMPUTES GAMMA FUNCTION OF A RATIONAL ARGUMENT,
USING AN O(T**2) METHOD, SPACE = 6T+12
CALL MPGAMQ (I, J, X) MEANS X = GAMMA(I/J)
MPGCD DIVIDES TWO INTEGERS BY THEIR GREATEST COMMON DIVISOR
(CALLED BY MPMULQ, MPGAMQ, ETC)
MPHANK ROUTINE CALLED BY MPBESJ (EVALUATES HANKELS ASYMPTOTIC
SERIES FOR BESSEL FUNCTIONS), SPACE = 11T+24
* MPIN CONVERTS FIXED-POINT NUMBER READ UNDER A1 FORMAT
TO MULTIPLE-PRECISION, SPACE = 3T+11
* MPINE SAME AS MPIN BUT RESULT IS MULTIPLIED BY A POWER OF
TEN (USEFUL FOR READING IN FLOATING-POINT NUMBERS),
SPACE = 5T+12
* MPLI EVALUATES LOGARITHMIC INTEGRAL LI(X), SPACE = 19T+31
CALL MPLI (X, Y) MEANS Y = LI(X)
* MPLN COMPUTES NATURAL LOG OF AN MP NUMBER,
USING AN O(SQRT(T).M(T)) METHOD, SPACE = 6T+14
CALL MPLN (X, Y) MEANS Y = LN(X)
MPLNGM COMPUTES LN(GAMMA(X)) FOR POSITIVE MP X,
USING STIRLINGS APPROXIMATION,
SPACE = 11T+24+NL*((T+3)/2), WHERE NL IS THE NUMBER
OF TERMS USED IN THE ASYMPTOTIC EXPANSION,
NL .LE. (2 + T*LN(B)/8)
CALL MPLNGM (X, Y) MEANS Y = LN(GAMMA(X))
MPLNGS COMPUTES NATURAL LOG OF AN MP NUMBER, USING
THE GAUSS-SALAMIN ALGORITHM. RECOMMENDED FOR
TESTING MPLN AND MPLNI ONLY (UNLESS T LARGE).
SPACE = 6T+26
* MPLNI COMPUTES NATURAL LOG OF AN INTEGER, USING AN
O(T**2) METHOD (FASTER THAN MPLN), SPACE = 3T+8
CALL MPLNI (N, X) MEANS X = LN(N)
MPLNS COMPUTES LN(1+X) FOR SMALL MP X, SPACE = 5T+12
MPL235 COMPUTES NATURAL LOG OF AN INTEGER WHOSE PRIME
FACTORS ARE 2, 3 AND/OR 5 (CALLED BY MPLNI),
SPACE = 3T+8
* MPMAX COMPUTES THE MAXIMUM OF TWO MP NUMBERS
CALL MPMAX (X, Y, Z) MEANS Z = MAX(X,Y)
* MPMAXR COMPUTES THE LARGEST POSITIVE MP NUMBER
CALL MPMAXR (X) MEANS X = MP NUMBER WITH EXPONENT M
AND ALL DIGITS B-1
* MPMIN COMPUTES THE MINIMUM OF TWO MP NUMBERS
CALL MPMIN (X, Y, Z) MEANS Z = MIN(X,Y)
* MPMINR RETURNS THE SMALLEST NORMALIZED POSITIVE MP NUMBER
CALL MPMINR (X) MEANS X = B**(-M-1)
MPMLP INNER LOOP ROUTINE CALLED BY MPMUL
* MPMUL MULTIPLIES TWO MP NUMBERS
USING AN M(T) = O(T**2) ALGORITHM
CALL MPMUL (X, Y, Z) MEANS Z = X*Y
* MPMULI MULTIPLIES AN MP NUMBER BY AN
INTEGER USING AN O(T) METHOD (FASTER THAN MPMUL)
CALL MPMULI (X, IY, Z) MEANS Z = X*IY
* MPMULQ MULTIPLIES MP NUMBER BY A RATIONAL NUMBER
CALL MPMULQ (X, I, J, Y) MEANS Y = X*I/J
MPMUL2 ROUTINE CALLED BY MPMULI
* MPNEG REVERSES SIGN OF AN MP NUMBER
CALL MPNEG (X, Y) MEANS Y = -X
MPNZR NORMALIZES AND ROUNDS OR TRUNCATES (CALLED BY
MPADD2, MPDIVI, MPMUL AND MPMUL2)
* MPOUT CONVERTS MULTIPLE-PRECISION TO A FORM SUITABLE FOR
PRINTING UNDER A1 FORMAT (CORRESPONDS TO F OR I
FORMATS), SPACE = 3T+11
* MPOUTE SIMILAR TO MPOUT BUT GIVES (DECIMAL) EXPONENT AND
FRACTION (CORRESPONDS TO E FORMAT), SPACE = 6T+14
MPOUT2 SAME AS MPOUT BUT ANY BASE FROM 2 TO 16 MAY BE
USED FOR OUTPUT REPRESENTATION, SPACE = 3T+11
MPOVFL ROUTINE CALLED ON MULTIPLE-PRECISION OVERFLOW
(CALLS MPERR AT PRESENT BUT EASILY MODIFIED)
* MPPACK PACKS MP NUMBERS INTO ARRAYS OF DIMENSION
(T+3)/2 (USEFUL TO SAVE SPACE),
UNPACKING MAY BE PERFORMED WITH MPUNPK
* MPPI RETURNS PI TO MULTIPLE-PRECISION ACCURACY,
USING AN O(T**2) METHOD, SPACE = 3T+8
CALL MPPI (X) MEANS X = 3.14159265...
MPPIGL RETURNS PI TO MULTIPLE-PRECISION ACCURACY,
USING GAUSS-LEGENDRE O(LOG(T)M(T)) METHOD,
RECOMMENDED FOR TESTING MPPI ONLY, SPACE = 6T+14
* MPPOLY EVALUATES A POLYNOMIAL WITH INTEGER COEFFICIENTS,
SPACE = 3T+8
* MPPWR RAISES MP NUMBER TO INTEGER POWER,
SPACE = 4T+10
CALL MPPWR (X, N, Y) MEANS Y = X**N
* MPPWR2 RAISES NONNEGATIVE MP NUMBER TO MP POWER,
SPACE = 7T+16
CALL MPPWR2 (X, Y, Z) MEANS Z = X**Y
* MPQPWR RAISES RATIONAL NUMBER TO RATIONAL POWER,
SPACE = 4T+10
CALL MPQPWR (I, J, K, L, X) MEANS X = (I/J)**(K/L)
* MPREC FORMS RECIPROCAL OF MP NUMBER,
USING NEWTONS METHOD, SPACE = 4T+10
CALL MPREC (X, Y) MEANS Y = 1/X
* MPROOT COMPUTES THE N-TH ROOT OF AN MP NUMBER
USING NEWTONS METHOD, SPACE = 4T+10
CALL MPROOT (X, N, Y) MEANS Y = X**(1/N)
* MPSET SETS THE BASE B AND DIGITS T ETC GIVEN THE
EQUIVALENT NUMBER OF DECIMAL PLACES REQUIRED
WARNING - MAY CAUSE AN INTEGER OVERFLOW,
******* FOR DETAILS SEE COMMENTS IN MPSET
* MPSIN COMPUTES SINE OF AN MP NUMBER,
USING AN O(M(T)T/LOG(T)) METHOD, SPACE = 5T+12
CALL MPSIN (X, Y) MEANS Y = SIN(X)
* MPSINH COMPUTES HYPERBOLIC SINE OF AN MP NUMBER,
USING MPEXP, SPACE = 5T+12
CALL MPSINH (X, Y) MEANS Y = SINH(X)
MPSIN1 COMPUTES SIN(X) OR COS(X) FOR ABS(X) .LE. 1, CALLED
BY MPSIN, MPCOS AND MPTAN, SPACE = 3T+8
* MPSQRT COMPUTES SQUARE ROOT OF A NONNEGATIVE MP NUMBER,
USING NEWTONS METHOD, SPACE = 4T+10
CALL MPSQRT (X, Y) MEANS Y = SQRT(X)
* MPSTR STORES ONE MP NUMBER IN ANOTHER
CALL MPSTR (X, Y) MEANS Y = X
* MPSUB SUBTRACTS ONE MP NUMBER FROM ANOTHER
CALL MPSUB (X, Y, Z) MEANS Z = X - Y
* MPTAN COMPUTES TAN OF AN MP NUMBER,
USING MPSIN1, SPACE = 6T+20
CALL MPTAN (X, Y) MEANS Y = TAN(X)
* MPTANH COMPUTES HYPERBOLIC TAN OF AN MP NUMBER,
USING MPEXP, SPACE = 5T+12
CALL MPTANH (X, Y) MEANS Y = TANH(X)
MPUNFL ROUTINE CALLED ON MULTIPLE-PRECISION UNDERFLOW
(SETS RESULT TO ZERO AT PRESENT BUT EASILY MODIFIED)
* MPUNPK UNPACKS AN ARRAY FORMED BY MPPACK TO GIVE AN MP
NUMBER IN STANDARD FORMAT
* MPZETA COMPUTES RIEMANN ZETA FUNCTION FOR POSITIVE
INTEGER ARGUMENTS
SPACE = 8T+18+NL*((T+3)/2), WHERE NL IS THE
NUMBER OF TERMS USED IN THE ASYMPTOTIC
EXPANSION, NL .LE. (1 + 0.1*T*LN(B))
CALL MPZETA (N, X) MEANS X = ZETA(N)
* MP40D OUTPUT ROUTINE CALLED BY TEST PROGRAM,
USEFUL FOR EASY FIXED-POINT OUTPUT,
SPACE = 3T+N+14 FOR N DECIMAL PLACE OUTPUT
CALL MP40D (N, X) WRITES X TO N DECIMAL PLACES ON UNIT
LUN, ASSUMING ABS(X) .LT. 10
MP40E OUTPUT ROUTINE CALLED BY MP40D
* MP40F OUTPUT ROUTINE CALLED BY TEST2 PROGRAM,
USEFUL FOR EASY FLOATING-POINT OUTPUT,
SPACE = 6T+N+17 FOR N SIGNIFICANT FIGURE OUTPUT
CALL MP40F (N, X) WRITES X TO N SIGNIFICANT FIGURES
(IN DECIMAL EXPONENT AND FRACTION FORM) ON UNIT LUN
MP40G OUTPUT ROUTINE CALLED BY MP40F
* TEST A MAIN PROGRAM WHICH TESTS SOME OF THE ROUTINES IN MP
WHILE COMPUTING VARIOUS CONSTANTS TO 40 DECIMAL PLACES
TESTV A VERSION OF TEST WITH VARIABLE-PRECISION COMPUTATION
AND OUTPUT
* TEST2 ANOTHER TEST PROGRAM WHICH TESTS ROUTINES
NOT CALLED BY TEST OR TESTV
TIMEMP A MACHINE-DEPENDENT FUNCTION CALLED BY TESTV, SHOULD
BE MODIFIED BY THE USER BEFORE TESTV IS RUN.
|
| This is taken from the list of large integer arithmetic
packages which I mentioned in the previous reply:
>bmp (Brent's Multiple Precision?)
> R. P. Brent
>
> 1981 vintage FORTRAN code to do extended precision floating &
> fixed point arithmetic. Includes most of the mathematical
> functions you'd find in a FORTRAN run-time library.
> This code is an ACM algorithm, number 524.
> To obtain, send a mail message to [email protected]
> containing the line "send mp.f from bmp" or better yet, perhaps
> just start with "help".
>[...]
>Mark Riordan [email protected]
>Michigan State University 19 April 1991
Everything after my "signature" is from the help message
from [email protected]. You may wish to start by sending
the one line message
send index from bmp
to decwrl::"[email protected]" to see what is there.
Dan
Return-Path: [email protected]
Received: by zfc.lmo.dec.com (5.57/fma-100391);
id AA23351; Tue, 16 Jun 92 16:34:16 -0400
Received: by inet-gw-1.pa.dec.com; id AA29643; Tue, 16 Jun 92 13:33:21 -0700
Received: by surfer.EPM.ORNL.GOV (5.61/1.34)id AA25427; Tue, 16 Jun 92 16:23:36 -0400
Date: Tue, 16 Jun 92 16:23:36 -0400
From: [email protected] (Netlib)
Message-Id: <[email protected]>
To: deramo
Subject: Re: Subject: help
===== How to use netlib =====
This file is the reply you'll get to:
mail [email protected]
send index
Here are examples of the various kinds of requests.
* get the full index for a library
send index from eispack
* get a particular routine and all it depends on
send dgeco from linpack
* get just the one routine, not subsidiaries
send only dgeco from linpack
* get dependency tree, but excluding a subtree
send dgeco but not dgefa from linpack
* just tell how large a reply would be, don't actually send the file
send list of dgeco from linpack
* get a list of sizes and times of all files in a library
send directory for benchmark
* search for somebody in the SIAM membership list:
who is gene golub
* keyword search for netlib software
find cubic spline
* bibliographic search
find schumaker from approximation
find aasen from linalg
* set the chunk size used for reply
mailsize 100k
* (optional) end of request
quit
The Internet address "[email protected]" refers to a gateway
machine, at Oak Ridge National Laboratory in Oak Ridge, Tennessee.
This address should be understood on all the major networks.
For access from Europe, try the duplicate collection in Oslo:
Internet: [email protected]
EARN/BITNET: netlib%[email protected]
X.400: s=netlib; o=nac; c=no;
EUNET/uucp: nac!netlib
For the Pacific, try [email protected]
located at the University of Wollongong, NSW, Australia.
A similar collection of statistical software is available from
[email protected].
The TeX User Group distributes TeX-related software from
[email protected].
The symbolic algebra system REDUCE is supported by
[email protected].
An excellent guide to the mysteries of networks and address syntax is:
Donnalyn Frey and Rick Adams (1989) "!%@:: A Directory of Electronic
Mail Addressing and Networks", O'Reilly & Associates, Inc, 632 Petaluma
Ave, Sebastopol CA 95472. Background about netlib is in Jack J.
Dongarra and Eric Grosse, Distribution of Mathematical Software Via
Electronic Mail, Comm. ACM (1987) 30,403-407 and in a quarterly column
published in the SIAM News and SIGNUM Newsletter.
Bugs reports, comments, and annual lists of recipients will be
forwarded to the code authors when possible. Many of these codes are
designed for use by professional numerical analysts who are capable of
checking for themselves whether an algorithm is suitable for their
needs. One routine can be superb and the next awful. So be careful!
An inventory list is given below and in the indices for the individual
libraries. If you know exactly what you're looking for, these guides
may be enough. An interactive system called "walk" provides a more
systematic list (not limited to netlib) but at present covers only
approximation. Volunteers from other fields are needed. The reference
is Eric Grosse, "A Catalog ...", in Algorithms for Approximation",
Mason and Cox (eds.), Chapman and Hall, 1989. Dialup (at 1200bps)
908-582-1238 or telnet to research.att.com and login as walk; no
password is required.
-------quick summary of contents---------
a - approximation algorithms
alliant - set of programs collected from Alliant users
amos - special functions by D. Amos. = toms/644
apollo - set of programs collected from Apollo users
benchmark - various benchmark programs and a summary of timings
bib - bibliographies
bihar - Bjorstad's biharmonic solver
bmp - Brent's multiple precision package
c - another "misc" library, for software written in C
cheney-kincaid - programs from the 1985 text
conformal - conformal mapping
contin - continuation, limit points
core - machine constants, vector and matrix * vector BLAS
c++ - code in the C++ language
dierckx - Spline fitting on various geometries.
domino - communication and scheduling of multiple tasks; Univ. Maryland
eispack - matrix eigenvalues and vectors
elefunt - Cody and Waite's tests for elementary functions
errata - corrections to numerical books
f2c - Fortran to C converter
fishpack - separable elliptic PDEs; Swarztrauber and Sweet
fitpack - Cline's splines under tension
fftpack - Swarztrauber's Fourier transforms
fmm - software from the book by Forsythe, Malcolm, and Moler
fn - Fullerton's special functions
fortran - single-double precision converter, static debugger
fp - floating point arithmetic
gcv - Generalized Cross Validation
gmat - multi-processing Time Line and State Graph tools, Mark Seager
go - "golden oldies" gaussq, zeroin, lowess, ...
graphics - auto color, ray-tracing benchmark
harwell - MA28 sparse linear system
hence - Heterogeneous Network Computing Environment
hompack - nonlinear equations by homotopy method
ieeecss - IEEE / Control Systems Society
itpack - iterative linear system solution by Young and Kincaid
jakef - automatic differentiation of Fortran subroutines
kincaid-cheney - programs from the 1990 text
lapack - solving the most common problems in numerical linear algebra
lanczos - Cullum and Willoughby's Lanczos programs
lanz - Large Sparse Symmetric Generalized Eigenproblem, Jones and Patrick
laso - Scott's Lanczos program for eigenvalues of sparse matrices
linpack - gaussian elimination, QR, SVD by Dongarra, Bunch, Moler, Stewart
lp - linear programming
machines - short descriptions of various computers
matlab - software from the MATLAB user's group
microscope - Alfeld and Harris' system for discontinuity checking
minpack - nonlinear equations and least squares by More, Garbow, Hillstrom
misc - everything else
ml - Standard ML of New Jersey (programming language compiler)
na-digest - archive of mailings to NA distribution list
napack - numerical algebra programs
news - Grosse's Netlib News column for na-net, SIAM News, SIGNUM Newsletter
numeralgo - algorithms from the new journal "Numerical Algorithms"
ode - ordinary differential equations
odepack - ordinary differential equations from Hindmarsh
odrpack - orthogonal distance regression, Boggs Byrd Donaldson Schnabel
opt - optimization
paragraph - display of algorithms on message-passing multiprocessor
paranoia - Kahan's floating point test
parmacs - parallel programmming macros
pascal - another "misc" library, for software written in Pascal
pchip - hermite cubics Fritsch+Carlson
pdes/madpack - a multigrid package, by Craig Douglas
picl - portable instrumented communication library for multiprocessors
pltmg - Bank's multigrid code; too large for ordinary mail
polyhedra - Hume's database of geometric solids
popi - Digital Darkroom image manipulation software (Holzmann)
port - the public subset of PORT library
posix - draft standards
pppack - subroutines from de Boor's Practical Guide to Splines
pvm - parallel virtual machine
quadpack - univariate quadrature by Piessens, de Donker, Kahaner
research - miscellanea from AT&T Bell Labs, Computing Science Research Center
sched - environment for portable parallel algorithms in a Fortran setting.
sciport - portable version of Cray SCILIB, by McBride and Lamson
sequent - software from the Sequent Users Group
slap - Seager + Greenbaum, iterative methods for symmetric and unsymmetric
slatec - error handling package from the Slatec library
sparse - Kundert + Sangiovanni-Vincentelli, C sparse linear algebra
sparse-blas - BLAS by indirection
sparspak - George + Liu, sparse linear algebra core
specfun - transportable special functions
spin - simulation and validation of communication protocols, G. Holzmann
stringsearch - string matching
toeplitz - linear systems in Toeplitz or circulant form by Garbow
toms - Collected Algorithms of the ACM
typesetting - typesetting macros and preprocessors
uncon/data - optimization test problems
vanhuffel - total least squares, partial SVD by Van Hufell
vfftpk - vectorized FFT; variant of fftpack
voronoi - Voronoi diagrams and Delaunay triangulations
y12m - sparse linear system (Aarhus)
--------a bit more detail--------
The first few libraries here are widely regarded as being of high quality.
The likelihood of your encountering a bug is relatively small; if you do,
we certainly want to hear about it! mail [email protected]
[I deleted the stuff after this point, except for the "bmp" item.--Dan]
lib bmp
for Brent's multiple precision package
master research.att.com
|