October 2003 - PPL-devel - BUGSENG Mailman

Re: Question about PPL (missing detail)
by Roberto Bagnara 08 Oct '03

08 Oct '03

Carlos Ureña Almagro wrote: > I'm trying to use PPL library for a project related to Computer > Graphics. First at all, I would like to thank you for your effort > in development and maintenence of that library. I had a trouble > trying to use it, and I would like to know if you have experienced > this and maybe you can point me to the solution. I apologize if > this is a trivial question, or has a trivial solution which I > ignore. > > I have suse linux 8.2, with gcc version 3.2, kernel 2.4.19. I > have succesfully installed the binary rpm version of your library. > (I have libstdc++.so.5 in my linux box, rpm missed the corresponding > libstdc++ rpm package, but the file is there, so I ignored the rpm > warning). > > After installing, I checked I had the libraries in standard linux lib > folders, and I wrote a simple C++ program using the PPL. The > program compiled OK, but when I tried to link it, I had an error > message, more concretelly, the linker found references to this > undefined symbol (dots are mine) > > standard_alloc_template<...>::_S_force_new > > may be the libraries I have are wrong, but every symbol was > correctly resolved except for this, and this is what bothers > me > > > Any help would be appreciated, thank you very much in advance. Dear Carlos, the symptom suggests that you are trying to link C++ objects obtained with different version of GCC, something that does not work in general. Things will improve in the future, but until now the situation has been quite discouraging: version 3.0 was not compatible with 3.1 and the same thing happened in the transition to 3.2 and to 3.3. Indeed, even the fact that we distribute RPMs is questionable, since they are useless on systems where a different version of GCC is installed. The problem is aggravated by the fact explained in our draft FAQ available at http://www.cs.unipr.it/ppl/FAQ. The bottom line is that GMP, the PPL and all your C++ code must be compiled with the same version of the same compiler (GCC, in your case). To summarize: you should build everything from sources. Start with GMP following the instructions in http://www.cs.unipr.it/ppl/Download/requirements and then configure, compile and install the PPL. Please do not hesitate to contact us again if you need help: direct all communication to ppl-devel(a)cs.unipr.it. We would also love to know more about your application. All the best, Roberto and the PPL team -- Prof. Roberto Bagnara Computer Science Group Department of Mathematics, University of Parma, Italy http://www.cs.unipr.it/~bagnara/ mailto:bagnara@cs.unipr.it

1 0

[Fwd: Question about PPL (missing detail)]
by Roberto Bagnara 08 Oct '03

08 Oct '03

-------- Original Message -------- Subject: Question about PPL (missing detail) Date: Wed, 08 Oct 2003 13:15:43 +0200 From: Carlos Ureña Almagro <almagro(a)ugr.es> Organization: ETSI Informática - Univ. Granada To: bagnara(a)cs.unipr.it Hello again, In my previous message, I forgott to mention that I was using the last version available of ppl when I obtained that linker error. thanks again, Carlos. -------- Original Message -------- Subject: Question about PPL Date: Wed, 08 Oct 2003 12:49:30 +0200 From: Carlos Ureña Almagro <almagro(a)ugr.es> Organization: ETSI Informática - Univ. Granada To: bagnara(a)cs.unipr.it Hello, I'm trying to use PPL library for a project related to Computer Graphics. First at all, I would like to thank you for your effort in development and maintenence of that library. I had a trouble trying to use it, and I would like to know if you have experienced this and maybe you can point me to the solution. I apologize if this is a trivial question, or has a trivial solution which I ignore. I have suse linux 8.2, with gcc version 3.2, kernel 2.4.19. I have succesfully installed the binary rpm version of your library. (I have libstdc++.so.5 in my linux box, rpm missed the corresponding libstdc++ rpm package, but the file is there, so I ignored the rpm warning). After installing, I checked I had the libraries in standard linux lib folders, and I wrote a simple C++ program using the PPL. The program compiled OK, but when I tried to link it, I had an error message, more concretelly, the linker found references to this undefined symbol (dots are mine) standard_alloc_template<...>::_S_force_new may be the libraries I have are wrong, but every symbol was correctly resolved except for this, and this is what bothers me Any help would be appreciated, thank you very much in advance. ---------------------------------------------------------------------- Carlos Urena-Almagro (almagro(a)ugr.es - http://lsi.ugr.es/~curena) Dpt. Lenguajes y Sistemas Informaticos - University of Granada - Spain. -- ----------------------------------------------------------------- Carlos Urena-Almagro (almagro(a)ugr.es - http://lsi.ugr.es/~curena) Dpt. Lenguajes y Sistemas Informaticos - University of Granada - Spain. -- Prof. Roberto Bagnara Computer Science Group Department of Mathematics, University of Parma, Italy http://www.cs.unipr.it/~bagnara/ mailto:bagnara@cs.unipr.it

1 0

[Fwd: Question about PPL]
by Roberto Bagnara 08 Oct '03

08 Oct '03

-------- Original Message -------- Subject: Question about PPL Date: Wed, 08 Oct 2003 12:49:30 +0200 From: Carlos Ureña Almagro <almagro(a)ugr.es> Organization: ETSI Informática - Univ. Granada To: bagnara(a)cs.unipr.it Hello, I'm trying to use PPL library for a project related to Computer Graphics. First at all, I would like to thank you for your effort in development and maintenence of that library. I had a trouble trying to use it, and I would like to know if you have experienced this and maybe you can point me to the solution. I apologize if this is a trivial question, or has a trivial solution which I ignore. I have suse linux 8.2, with gcc version 3.2, kernel 2.4.19. I have succesfully installed the binary rpm version of your library. (I have libstdc++.so.5 in my linux box, rpm missed the corresponding libstdc++ rpm package, but the file is there, so I ignored the rpm warning). After installing, I checked I had the libraries in standard linux lib folders, and I wrote a simple C++ program using the PPL. The program compiled OK, but when I tried to link it, I had an error message, more concretelly, the linker found references to this undefined symbol (dots are mine) standard_alloc_template<...>::_S_force_new may be the libraries I have are wrong, but every symbol was correctly resolved except for this, and this is what bothers me Any help would be appreciated, thank you very much in advance. ---------------------------------------------------------------------- Carlos Urena-Almagro (almagro(a)ugr.es - http://lsi.ugr.es/~curena) Dpt. Lenguajes y Sistemas Informaticos - University of Granada - Spain. -- Prof. Roberto Bagnara Computer Science Group Department of Mathematics, University of Parma, Italy http://www.cs.unipr.it/~bagnara/ mailto:bagnara@cs.unipr.it

1 0

[Fwd: Re: Representing +inf, -inf and nan in mpq_class objects]
by Roberto Bagnara 07 Oct '03

07 Oct '03

-------- Original Message -------- Subject: Re: Representing +inf, -inf and nan in mpq_class objects Date: Tue, 07 Oct 2003 09:20:04 +1000 From: Kevin Ryde <user42(a)zip.com.au> To: Roberto Bagnara <bagnara(a)cs.unipr.it> CC: gmp-discuss(a)swox.com References: <3F7E76E5.1060100(a)cs.unipr.it> Roberto Bagnara <bagnara(a)cs.unipr.it> writes: > > resorting to normal mpz_class > operations only for normal values. Provided your special values don't reach the mpq functions there shouldn't be any problems. A few mpq functions will be happy with any values actually, eg. mpq_set and the input/output routines I think. > e.g., if the (present or future) implementation > does too much automatic canonicalization. There's no plans for anything like that. -- Prof. Roberto Bagnara Computer Science Group Department of Mathematics, University of Parma, Italy http://www.cs.unipr.it/~bagnara/ mailto:bagnara@cs.unipr.it

1 0

[Fwd: Representing +inf, -inf and nan in mpq_class objects]
by Roberto Bagnara 07 Oct '03

07 Oct '03

-------- Original Message -------- Subject: Representing +inf, -inf and nan in mpq_class objects Date: Sat, 04 Oct 2003 09:29:41 +0200 From: Roberto Bagnara <bagnara(a)cs.unipr.it> Organization: Department of Mathematics, University of Parma, Italy To: gmp-discuss(a)swox.com I need a class of rationals extended with +inf, -inf and nan, and I also need to minimize memory occupation. I am wondering if it is safe to represent them with the fractions 1/0, -1/0 and 0/0 as objects of type mpq_class. All the required operations (basically addition and comparisons is all I need) would be redefined so as to test for the special cases first and handle them in the appropriate way, resorting to normal mpz_class operations only for normal values. So I guess the question is: is there something in the present (and, conceivably, future) design of mpz_class for which - sticking 1/0, -1/0 or 0/0 into an object and - testing for the presence of such special values in an object may cause wrong or undefined behavior? This may happen, e.g., if the (present or future) implementation does too much automatic canonicalization. What do you think? Cheers, Roberto -- Prof. Roberto Bagnara Computer Science Group Department of Mathematics, University of Parma, Italy http://www.cs.unipr.it/~bagnara/ mailto:bagnara@cs.unipr.it _______________________________________________ gmp-discuss mailing list gmp-discuss(a)swox.com http://swox.com/mailman/listinfo/gmp-discuss -- Prof. Roberto Bagnara Computer Science Group Department of Mathematics, University of Parma, Italy http://www.cs.unipr.it/~bagnara/ mailto:bagnara@cs.unipr.it

1 0

ppl ./TODO tests/h79widening6.cc
by Roberto Bagnara 06 Oct '03

06 Oct '03

CVSROOT: /cvs/ppl Module name: ppl Changes by: roberto(a)cs.unipr.it 2003-10-06 19:02:38 Modified files: . : TODO tests : h79widening6.cc Log message: h79widening6.cc temporarily fixed, but must be turned into a proper test. Patches: http://www.cs.unipr.it/cgi-bin/cvsweb.cgi/ppl/TODO.diff?cvsroot=ppl&r1=1.59… http://www.cs.unipr.it/cgi-bin/cvsweb.cgi/ppl/tests/h79widening6.cc.diff?cv…

1 0

ppl/src Row.cc
by Roberto Bagnara 04 Oct '03

04 Oct '03

CVSROOT: /cvs/ppl Module name: ppl Changes by: roberto(a)cs.unipr.it 2003-10-04 22:16:06 Modified files: src : Row.cc Log message: Revert previous change. Patches: http://www.cs.unipr.it/cgi-bin/cvsweb.cgi/ppl/src/Row.cc.diff?cvsroot=ppl&r…

1 0

ppl/src Row.cc
by Roberto Bagnara 04 Oct '03

04 Oct '03

CVSROOT: /cvs/ppl Module name: ppl Changes by: roberto(a)cs.unipr.it 2003-10-04 22:07:08 Modified files: src : Row.cc Log message: Avoid including <iostream>. Patches: http://www.cs.unipr.it/cgi-bin/cvsweb.cgi/ppl/src/Row.cc.diff?cvsroot=ppl&r…

1 0

ppl ./.cvsignore Watchdog/.cvsignore
by Roberto Bagnara 03 Oct '03

03 Oct '03

CVSROOT: /cvs/ppl Module name: ppl Changes by: roberto(a)cs.unipr.it 2003-10-03 12:22:50 Modified files: . : .cvsignore Watchdog : .cvsignore Log message: Ignore config.h.in. Patches: http://www.cs.unipr.it/cgi-bin/cvsweb.cgi/ppl/.cvsignore.diff?cvsroot=ppl&r… http://www.cs.unipr.it/cgi-bin/cvsweb.cgi/ppl/Watchdog/.cvsignore.diff?cvsr…

1 0

ppl ./config.h.in Watchdog/config.h.in
by Roberto Bagnara 03 Oct '03

03 Oct '03

CVSROOT: /cvs/ppl Module name: ppl Changes by: roberto(a)cs.unipr.it 2003-10-03 09:44:57 Removed files: . : config.h.in Watchdog : config.h.in Log message: These should not be in the repository, since they are automatically generated. Patches: http://www.cs.unipr.it/cgi-bin/cvsweb.cgi/ppl/config.h.in.diff?cvsroot=ppl&… http://www.cs.unipr.it/cgi-bin/cvsweb.cgi/ppl/Watchdog/config.h.in.diff?cvs…

1 0