- CLN-list - ginac.de

swapping numbers
by Dejan Jovanović 26 Aug '10

26 Aug '10

Hi, I'm writing an application with intense usage of integers and one of the operations that might have a performance impact is swapping two numbers. I don't know the internals of the CLN library and I see there is no swap operation in the headers. Could someone please elaborate how much stuff happens internally when I do the usual swap: void swap(cl_I& a, cl_I& b) { (1) cl_I temp; (2) temp = a; (3) a = b; (4) b = temp; (5) } My worst-case guess is: (1) a sequence of constructors (2) reconstruction of a (3) destruction of a, reconstruction of b (4) destruction of b, reconstruction of temp (5) destruction of temp Also, would massive amount of swaps lead to internal memory fragmentation due to construction/destruction? Thanks, Dejan

3 5

how to read string numbers at runtime into cl_I type
by Joshua Friedman 21 Jul '10

21 Jul '10

I have a text file like this: 16 8534483450280192 17 236164889075956560 18 -2445879764812789440 19 14227278615873217344 ... Integer paired with a very large integer. How do I read in from a file to store the big integer in a cl_I type. I tried reading it in as a string, and then I couldn't convert from a string to any of the types (cl_R, or cl_I). Eventually I'll need it to be a float, which I should be able to use cl_float() to convert from an exact type. Thanks Josh -- Joshua Friedman PhD CrownEagle(a)gmail.com http://www.math.sunysb.edu/~joshua

3 3

Compile with MSVC successful
by Jan Rheinlaender 21 Jul '10

21 Jul '10

I successfully compiled CLN-1.3.1 with Microsoft Visual C++. You can find an account of my adventures on http://sourceforge.net/apps/mediawiki/ooo-imath/index.php?title=Installatio… The benchmarks, examples and tests compile and both tests claim to pass successfully. Since I am not an expert programmer, I have no idea if my source code modifications are acceptable or "evil". Have fun, Jan

1 0

file input and cln number types
by Joshua Friedman 15 Jul '10

15 Jul '10

I have a text file like this: 16 8534483450280192 17 236164889075956560 18 -2445879764812789440 19 14227278615873217344 ... Integer paired with a very large integer. How do I read in from a file to store the big integer in a cl_I type. I tried reading it in as a string, and then I couldn't convert from a string to any of the types (cl_R, or cl_I). Eventually I'll need it to be a float, which I should be able to use cl_float() to convert from an exact type. Thanks Josh -- Joshua Friedman PhD CrownEagle(a)gmail.com http://www.math.sunysb.edu/~joshua

1 0

[PATCH 2/2] Configure: be more cross-compilation friendly.
by Alexei Sheplyakov 28 May '10

28 May '10

Rewrite CL_GMP_SET_UINTD so it works for cross-compilation too. --- m4/gmp.m4 | 67 ++++++++++++++++++++++++++++++++++++------------------------ 1 files changed, 40 insertions(+), 27 deletions(-) diff --git a/m4/gmp.m4 b/m4/gmp.m4 index e0e5ef3..455c200 100644 --- a/m4/gmp.m4 +++ b/m4/gmp.m4 @@ -36,33 +36,46 @@ cl_cv_new_libgmp="yes", cl_cv_new_libgmp="no") dnl What is sizeof(mp_limb_t)? (It has to match sizeof(uintD) later.) AC_DEFUN([CL_GMP_SET_UINTD], [AC_CACHE_CHECK([how large gmp demands uintD to be], cl_cv_gmp_set_uintd, [ - AC_TRY_RUN([#include <gmp.h> -#include <stdio.h> -int main() { - FILE *f=fopen("conftestval", "w"); - if (!f) return(255); - if (sizeof(mp_limb_t) > sizeof(long)) - fprintf(f, "long long"); - else if (sizeof(mp_limb_t) == sizeof(long)) - fprintf(f, "long"); - else if (sizeof(mp_limb_t) == sizeof(int)) - fprintf(f, "int"); - else return(sizeof(mp_limb_t)); -#if defined(__GMP_BITS_PER_MP_LIMB) - /* Is there a nail in a limb? */ - if (8*sizeof(mp_limb_t)!=__GMP_BITS_PER_MP_LIMB) - return(254); -#endif - return(0); -}], cl_cv_gmp_set_uintd=`cat conftestval` - cl_gmp_demands="GMP_DEMANDS_UINTD_`echo ${cl_cv_gmp_set_uintd} | sed -e 'y/ gilnot/_GILNOT/'`", - gmp_retval="$ac_status" - if test x$gmp_retval = "x255"; then AC_MSG_ERROR([error opening output file.]); fi - if test x$gmp_retval = "x254"; then AC_MSG_ERROR([nails in MP limbs are unsupported.]); fi - AC_MSG_ERROR([Don't know which C-type has sizeof $gmp_retval.]), - AC_MSG_ERROR([cross-compiling - cannot determine])) -]) -AC_DEFINE_UNQUOTED($cl_gmp_demands) + dnl Note: we don't run any of compiled programs here, so this method + dnl both works for native and cross compilation + cl_gmp_demands="UNKNOWN" + cl_gmp_has_nails="no" + AC_COMPILE_IFELSE([AC_LANG_PROGRAM([[#include <gmp.h> + template<bool COND> struct Static_Assert; + template<> struct Static_Assert<true> { }; + #if defined(__GMP_BITS_PER_MP_LIMB) + Static_Assert<8*sizeof(mp_limb_t) == __GMP_BITS_PER_MP_LIMB> check; + #endif]], [[]])], [], [cl_gmp_has_nails="yes"]) + if test "x$cl_gmp_has_nails" = "xyes"; then + AC_MSG_ERROR([nails in MP libms are unsupported.]) + fi + AC_COMPILE_IFELSE([AC_LANG_PROGRAM([[#include <gmp.h> + template<bool COND> struct Static_Assert; + template<> struct Static_Assert<true> { }; + Static_Assert<sizeof(mp_limb_t) > sizeof(long)> check;]], [[]])], + [cl_gmp_demands='GMP_DEMANDS_UINTD_LONG_LONG'], []) + if test "x$cl_gmp_demands" = "xUNKNOWN"; then + AC_COMPILE_IFELSE([AC_LANG_PROGRAM([[#include <gmp.h> + template<bool COND> struct Static_Assert; + template<> struct Static_Assert<true> { }; + Static_Assert<sizeof(mp_limb_t) == sizeof(long)> check;]], [[]])], + [cl_gmp_demands='GMP_DEMANDS_UINTD_LONG'], []) + fi + if test "x$cl_gmp_demands" = "xUNKNOWN"; then + AC_COMPILE_IFELSE([AC_LANG_PROGRAM([[ + #include <gmp.h> + template<bool COND> struct Static_Assert; + template<> struct Static_Assert<true> { }; + Static_Assert<sizeof(mp_limb_t) == sizeof(int)> check;]], [[]])], + [cl_gmp_demands='GMP_DEMANDS_UINTD_INT'], []) + fi + if test "x$cl_gmp_demands" = "xUNKNOWN"; then + AC_MSG_ERROR([Don't know which C-type has sizeof(mp_limb_t)]) + else + cl_cv_gmp_set_uintd="$cl_gmp_demands" + fi + ]) +AC_DEFINE_UNQUOTED($cl_cv_gmp_set_uintd) ]) dnl Whether or not to use GMP. Sets CL_USE_GMP. -- 1.7.1

1 0

[PATCH 1/2] Fix the documentation build failure (wrong texinfo).
by Alexei Sheplyakov 28 May '10

28 May '10

Building PDF, PostScript, and DVI versions of the documentation fails with the following error message(s): ../../doc/cln.texi:1170: Argument of @asis has an extra }. <inserted text> @par <to be read again> } According to texinfo maintainer @itemize @asis is illegal, one should use @itemize @w{} instead [1]. [1] http://lists.gnu.org/archive/html/bug-texinfo/2009-03/msg00016.html --- doc/cln.texi | 12 ++++++------ 1 files changed, 6 insertions(+), 6 deletions(-) diff --git a/doc/cln.texi b/doc/cln.texi index 7c0aa5f..d16f8af 100644 --- a/doc/cln.texi +++ b/doc/cln.texi @@ -1167,7 +1167,7 @@ Returns the complex conjugate of @code{x}. We have the relations -@itemize @asis +@itemize @w{} @item @code{x = complex(realpart(x), imagpart(x))} @item @@ -1282,7 +1282,7 @@ down exactly as often as it rounds up. The functions are related like this: -@itemize @asis +@itemize @w{} @item @code{ceiling(m/n) = floor((m+n-1)/n) = floor((m-1)/n)+1} for rational numbers @code{m/n} (@code{m}, @code{n} integers, @code{n}>0), and @@ -2091,7 +2091,7 @@ come from @code{newbyte} and the other bits come from @code{n}. The following relations hold: -@itemize @asis +@itemize @w{} @item @code{ldb (n, b) = mask_field(n, b) >> b.position}, @item @@ -2454,7 +2454,7 @@ If @code{x} is already an exact number, this function returns @code{x}. If @code{x} is any float, one has -@itemize @asis +@itemize @w{} @item @code{cl_float(rational(x),x) = x} @item @@ -2628,7 +2628,7 @@ one digit in the non-exponent part. The exponent has the syntax @var{expmarker} @var{expsign} @{@var{digit}@}+. The exponent marker is -@itemize @asis +@itemize @w{} @item @samp{s} for short-floats, @item @@ -2646,7 +2646,7 @@ function @samp{float_format}. @item Complex numbers External representation: -@itemize @asis +@itemize @w{} @item In algebraic notation: @code{@var{realpart}+@var{imagpart}i}. Of course, if @var{imagpart} is negative, its printed representation begins with -- 1.7.1

1 0

Re: [CLN-list] How to compile GiNaC under MinGW/MSYS (Alexei Sheplyakov)
by wang wayne 28 May '10

28 May '10

Thank you ,Alexei but that didn't work. i had tried, :( i suppose the point is how comes this message when compiling and linking cln : >C:/ooo/1.0/GiNaC1/lib/libcln.a(cl_LF_1plus.o): In function `subx_loop_up': >C:\msys\1.0\home\Lenovo\cln-1.3.1\src/./base/digitseq/cl_DS.h:406: undefined >reference to `_imp____gmpn_sub_n' it seems that the CLN and GINAC was compiled with position dependent things written into cln lib files , By the Way, when i type "pkg-config.exe --libs cln" in MSYS shell, it gives "-LC:/ooo/1.0/GiNaC1/lib -lcln" . it should have been "-L/GiNaC1/lib -lcln" , am i right? wayne

1 0

CLN compile errors on GCC4.2
by Vipin Varma 26 May '10

26 May '10

Hi, I've been using CLN 1.3.1 successfully with the GCC4.3 compiler. Now I am running my code using the GCC4.2 on another machine (in which I dont have the option of updating to the newer GCC): <SNIP> In file included from /users/varma/cln-1.3.1/include/cln/string.h:7, from /usr/include/c++/4.2/cstring:52, from /usr/include/c++/4.2/i486-linux-gnu/bits/c++locale.h:47, from /usr/include/c++/4.2/iosfwd:45, from /usr/include/c++/4.2/ios:43, from /usr/include/c++/4.2/ostream:45, from /usr/include/c++/4.2/iostream:45, from graph.cpp:12: /usr/include/cln/io.h:22: error: expected initializer before & token /usr/include/cln/io.h:23: error: expected initializer before & token /usr/include/cln/io.h:24: error: expected initializer before * token ........ ........ ........ //and many more errors// <SNIP> Lines 22-24 of cln/io.h contain the following: <SNIP> typedef std::istream& cl_istream; typedef std::ostream& cl_ostream; extern std::ostream* cl_debugout_stream; #define cl_debugout (*cl_debugout_stream) <SNIP> My guess is that the streams are not recognised due to circular referencing (but which are the two, or more, files that are responsibile for this, I am not sure), which is somehow taken care of in the newer GCC. Any advice on this will be appreciated. Vipin

3 6

How to compile GiNaC under MinGW/MSYS
by wang wayne 26 May '10

26 May '10

Hi,everyone. is there anyone here could help me, please ,i have built GiNaC dozens of times, now im in desperate. here are my information: 1) OS: windows 7 home basic MSYS 1.0.11 , TDM/MinGW 4.4.1 gmp5.0.1 , cln 1.3.1 ,GiNaC 1.5.7 2)building gmp (./configure --prefix=/GiNaC --enable-cxx --disable-static --enable-shared), normal downloading binary version of pkg-config and dll of glib from GNOME websites.( http://ftp.gnome.org/pub/gnome/binaries/win32/dependencies/pkg-config_0.23-…, http://ftp.gnome.org/pub/gnome/binaries/win32/glib/2.24/glib_2.24.1-1_win32… ) coping pkg-config.exe and libglib-2.0-0.dll to msys/bin. 3) building cln(./configure --prefix=/GiNaC --with-gmp=/GiNaC --disable-static --enable-shared ),but i only got the static libs 4)building GiNaC (./configure --prefix=/GiNaC --disable-static --enable-shared ) still only static libs 5)compiling example codes for testing. gmp OK,but cln gives me this error: g++ pi.cc -I/GiNaC1/include -L/GiNaC1/lib -lcln -o pi 2>a.txt C:/ooo/1.0/GiNaC1/lib/libcln.a(cl_LF_1plus.o): In function `subx_loop_up': C:\msys\1.0\home\Lenovo\cln-1.3.1\src/./base/digitseq/cl_DS.h:406: undefined reference to `_imp____gmpn_sub_n' C:/ooo/1.0/GiNaC1/lib/libcln.a(cl_LF_1plus.o): In function `shiftrightcopy_loop_down': C:\msys\1.0\home\Lenovo\cln-1.3.1\src/./base/digitseq/cl_DS.h:495: undefined reference to `_imp____gmpn_rshift' C:/ooo/1.0/GiNaC1/lib/libcln.a(cl_LF_1plus.o): In function `add_loop_up': C:\msys\1.0\home\Lenovo\cln-1.3.1\src/./base/digitseq/cl_DS.h:378: undefined reference to `_imp____gmpn_add_n' ............................................................. C:\msys\1.0\home\Lenovo\cln-1.3.1\src/./base/digitseq/cl_DS.h:453: undefined reference to `_imp____gmpn_lshift' C:/ooo/1.0/GiNaC1/lib/libcln.a(cl_I_to_digits.o): In function `divu_loop_down': C:\msys\1.0\home\Lenovo\cln-1.3.1\src/./base/digitseq/cl_DS.h:539: undefined reference to `_imp____gmpn_divrem_1' C:/ooo/1.0/GiNaC1/lib/libcln.a(cl_I_ldbx.o): In function `shiftrightcopy_loop_down': C:\msys\1.0\home\Lenovo\cln-1.3.1\src/./base/digitseq/cl_DS.h:495: undefined reference to `_imp____gmpn_rshift' collect2: ld returned 1 exit status

2 1

Re: [CLN-list] CLN-list Digest, Vol 49, Issue 3
by Joshua Friedman 26 May '10

26 May '10

Alexei Thanks for answering my question. I think I understand what to do. Suppose I say, cln_R x = 3.4; then I computer exp(x), now I convert it to cln_F, Can I do so at any precision. Is exp(x) just a symbolic computation, and does the actual computation come about only when converting to a cln_F? Thanks On Tue, May 18, 2010 at 4:45 PM, <cln-list-request(a)ginac.de> wrote: > Send CLN-list mailing list submissions to > cln-list(a)ginac.de > > To subscribe or unsubscribe via the World Wide Web, visit > https://www.cebix.net/mailman/listinfo/cln-list > or, via email, send a message with subject or body 'help' to > cln-list-request(a)ginac.de > > You can reach the person managing the list at > cln-list-owner(a)ginac.de > > When replying, please edit your Subject line so it is more specific > than "Re: Contents of CLN-list digest..." > > > Today's Topics: > > 1. confusion with cln::cl_F and cln::cl_R (Joshua Friedman) > 2. Re: confusion with cln::cl_F and cln::cl_R (Alexei Sheplyakov) > > > ---------------------------------------------------------------------- > > Message: 1 > Date: Tue, 18 May 2010 11:22:25 -0400 > From: Joshua Friedman <crowneagle(a)gmail.com> > To: cln-list(a)ginac.de > Subject: [CLN-list] confusion with cln::cl_F and cln::cl_R > Message-ID: > <AANLkTimibjqTFsn4dXLgX946oMrL7E7rmMaiu-eEZFVk(a)mail.gmail.com> > Content-Type: text/plain; charset="windows-1252" > > The program runs when I define "bigF = cln_R", but I get errors when I set > "bigF = cln_F" > I found that the exponential function only gives many digits of precision > when I use cln_F. > I am attaching my program and the out put is below: > > fuji:rama joshua$ make > g++ -DHAVE_CONFIG_H -I. -I../include -I../include -g -O2 -MT e.o -MD -MP > -MF .deps/e.Tpo -c -o e.o e.cc > In file included from /usr/include/c++/4.0.0/backward/iostream.h:31, > from e.cc:1: > /usr/include/c++/4.0.0/backward/backward_warning.h:32:2: warning: #warning > This file includes at least one deprecated or antiquated header. Please > consider using one of the 32 headers found in section 17.4.1.2 of the C++ > standard. Examples include substituting the <X> header for the <X.h> header > for C++ includes, or <iostream> instead of the deprecated header > <iostream.h>. To disable this warning use -Wno-deprecated. > e.cc: In function ?bigF h(bigF, bigF*, int, int, int)?: > e.cc:233: error: conversion from ?const cln::cl_N? to non-scalar type > ?bigF? > requested > e.cc: In function ?bigF ff(bigF, bigF, bigF*, int, int, int)?: > e.cc:251: error: no match for ?operator=? in ?q = cln::operator*(const > cln::cl_R&, const cln::cl_R&)(((const cln::cl_R&)(& cln::exp(const > cln::cl_R&)())))? > ../include/cln/float_class.h:40: note: candidates are: cln::cl_F& > cln::cl_F::operator=(const cln::cl_F&) > ../include/cln/float_class.h:47: note: cln::cl_F& > cln::cl_F::operator=(float) > ../include/cln/float_class.h:48: note: cln::cl_F& > cln::cl_F::operator=(double) > e.cc:258: error: conversion from ?const cln::cl_N? to non-scalar type > ?bigF? > requested > e.cc: In function ?bigF v(bigF, bigF*, int, int, int)?: > e.cc:275: error: conversion from ?int? to ?cln::cl_F? is ambiguous > ../include/cln/float_class.h:38: note: candidates are: > cln::cl_F::cl_F(cln::cl_anything*) > ../include/cln/float_class.h:26: note: > cln::cl_F::cl_F(const > char*) > ../include/cln/float_class.h:48: note: > cln::cl_F::cl_F(double) > ../include/cln/float_class.h:47: note: > cln::cl_F::cl_F(float) > e.cc:276: error: conversion from ?int? to ?cln::cl_F? is ambiguous > ../include/cln/float_class.h:38: note: candidates are: > cln::cl_F::cl_F(cln::cl_anything*) <near match> > ../include/cln/float_class.h:26: note: > cln::cl_F::cl_F(const > char*) <near match> > ../include/cln/float_class.h:48: note: > cln::cl_F::cl_F(double) > ../include/cln/float_class.h:47: note: > cln::cl_F::cl_F(float) > e.cc:281: error: no match for ?operator=? in ?w = cln::operator*(const > cln::cl_R&, const cln::cl_R&)(((const cln::cl_R&)((const cln::cl_R*)(& > w.cln::cl_F::<anonymous>))))? > ../include/cln/float_class.h:40: note: candidates are: cln::cl_F& > cln::cl_F::operator=(const cln::cl_F&) > ../include/cln/float_class.h:47: note: cln::cl_F& > cln::cl_F::operator=(float) > ../include/cln/float_class.h:48: note: cln::cl_F& > cln::cl_F::operator=(double) > e.cc:283: error: ambiguous overload for ?operator=? in ?q = 0? > ../include/cln/float_class.h:40: note: candidates are: cln::cl_F& > cln::cl_F::operator=(const cln::cl_F&) > ../include/cln/float_class.h:47: note: cln::cl_F& > cln::cl_F::operator=(float) > ../include/cln/float_class.h:48: note: cln::cl_F& > cln::cl_F::operator=(double) > e.cc:289: error: no match for ?operator=? in ?q = cln::operator*(const > cln::cl_R&, const cln::cl_R&)(((const cln::cl_R&)(& cln::exp(const > cln::cl_R&)())))? > ../include/cln/float_class.h:40: note: candidates are: cln::cl_F& > cln::cl_F::operator=(const cln::cl_F&) > ../include/cln/float_class.h:47: note: cln::cl_F& > cln::cl_F::operator=(float) > ../include/cln/float_class.h:48: note: cln::cl_F& > cln::cl_F::operator=(double) > e.cc:294: error: conversion from ?const cln::cl_N? to non-scalar type > ?bigF? > requested > e.cc: In function ?bigF integralH(bigF*, int, int, int, int)?: > e.cc:312: error: ambiguous overload for ?operator=? in ?aa = 1? > ../include/cln/float_class.h:40: note: candidates are: cln::cl_F& > cln::cl_F::operator=(const cln::cl_F&) > ../include/cln/float_class.h:47: note: cln::cl_F& > cln::cl_F::operator=(float) > ../include/cln/float_class.h:48: note: cln::cl_F& > cln::cl_F::operator=(double) > e.cc:313: error: ambiguous overload for ?operator=? in ?b = 20? > ../include/cln/float_class.h:40: note: candidates are: cln::cl_F& > cln::cl_F::operator=(const cln::cl_F&) > ../include/cln/float_class.h:47: note: cln::cl_F& > cln::cl_F::operator=(float) > ../include/cln/float_class.h:48: note: cln::cl_F& > cln::cl_F::operator=(double) > e.cc:317: error: conversion from ?int? to ?cln::cl_F? is ambiguous > ../include/cln/float_class.h:38: note: candidates are: > cln::cl_F::cl_F(cln::cl_anything*) > ../include/cln/float_class.h:26: note: > cln::cl_F::cl_F(const > char*) > ../include/cln/float_class.h:48: note: > cln::cl_F::cl_F(double) > ../include/cln/float_class.h:47: note: > cln::cl_F::cl_F(float) > e.cc:320: error: ambiguous overload for ?operator=? in ?w = 0? > ../include/cln/float_class.h:40: note: candidates are: cln::cl_F& > cln::cl_F::operator=(const cln::cl_F&) > ../include/cln/float_class.h:47: note: cln::cl_F& > cln::cl_F::operator=(float) > ../include/cln/float_class.h:48: note: cln::cl_F& > cln::cl_F::operator=(double) > e.cc:321: error: conversion from ?int? to ?cln::cl_F? is ambiguous > ../include/cln/float_class.h:38: note: candidates are: > cln::cl_F::cl_F(cln::cl_anything*) <near match> > ../include/cln/float_class.h:26: note: > cln::cl_F::cl_F(const > char*) <near match> > ../include/cln/float_class.h:48: note: > cln::cl_F::cl_F(double) > ../include/cln/float_class.h:47: note: > cln::cl_F::cl_F(float) > e.cc: In function ?bigF integralL(bigF*, int, int, int, int)?: > e.cc:339: error: ambiguous overload for ?operator=? in ?b = 1? > ../include/cln/float_class.h:40: note: candidates are: cln::cl_F& > cln::cl_F::operator=(const cln::cl_F&) > ../include/cln/float_class.h:47: note: cln::cl_F& > cln::cl_F::operator=(float) > ../include/cln/float_class.h:48: note: cln::cl_F& > cln::cl_F::operator=(double) > e.cc:346: error: ambiguous overload for ?operator=? in ?w = 0? > ../include/cln/float_class.h:40: note: candidates are: cln::cl_F& > cln::cl_F::operator=(const cln::cl_F&) > ../include/cln/float_class.h:47: note: cln::cl_F& > cln::cl_F::operator=(float) > ../include/cln/float_class.h:48: note: cln::cl_F& > cln::cl_F::operator=(double) > make: *** [e.o] Error 1 > fuji:rama joshua$ make > > > > > > -- > Joshua Friedman PhD > CrownEagle(a)gmail.com > http://www.math.sunysb.edu/~joshua <http://www.math.sunysb.edu/%7Ejoshua> >

2 1