On Tue, Oct 06, 2015 at 08:02:50AM -0400, Rafael de Farias Meurer via RT wrote:
Show quoted text> Hi, I'm trying to install Math::Pari in solaris 5.11 but i have errors, Can you help me?
>
> ERROR IN MAKE:
> /usr/perl5/5.8.4/bin/perlgcc Makefile.PL CCCDLFLAGS=-fPICSetting up Math::Pari with Perl 5.008004 on solaris version 2.11; cc=gcc, gccversion=, cccdlflags='-KPIC', ccflags='-D_LARGEFILE_SOURCE -D_FILE_OFFSET_BITS=64 -xarch=v8 -D_TS_ERRNO', optimize='-xO3 -xspace -xildoff', ld=gcc, ldflags=''.Found GP/PARI build directory in ./pari-2.1.7 (integer-formatted version 2001007).
> ### GP/PARI library version 2.1.* has many problems fixed in versions 2.3.*.
> ### Especially important are the problems with symbols `hiremainder' and ### `overflow' - if your build fails and error messages contain these ### words, updating to 2.3.* might allow your use of fast processor-specific ### assembler code.)
> ### Switching to 2.3.* should be as simple as giving Makefile.PL an option ### version23_ok=1 ### (if you or Makefile.PL already downloaded-and-extracted version-2.1.* ### distribution, give an extra option, force_download). ### Alternatively, give option ### paridir=DIRECTORY-of-extracted-distribution
> ### However, keep in mind: v.2.3.* supports only plotting to PostScript.
> ...Will not overwrite libPARI/paricfg.h... (remove it manually if needed) You may also want to remove libPARI/paricfg.h if your configuration changed from the time of the first build in this directory......Processor of family `sparcv9' detected...I will use assembler build of type 'sparcv8_micro'....Assembler is not GNU assemblerWriting Makefile for Math::PARI::libPARIWriting Makefile for Math::PariLooking for patches for 2001007...
> The following patches were judged to be beneficial, but were not applied:
> Suporte Especializado ROOT : makecd libPARI && make LIB="" LIBPERL_A="libperl.a" LINKTYPE="dynamic" PREFIX="" OPTIMIZE="-xO3 -xspace -xildoff" PASTHRU_DEFINE="" PASTHRU_INC=""gcc -c -I .././pari-2.1.7/src -I .././pari-2.1.7/src/headers -I .././pari-2.1.7/src/graph -I . -D_LARGEFILE_SOURCE -D_FILE_OFFSET_BITS=64 -xarch=v8 -D_TS_ERRNO -xO3 -xspace -xildoff -DVERSION=\"\" -DXS_VERSION=\"\" -KPIC -Derr=pari_err -DSOLARIS -DHAS_STAT -DHAS_OPENDIR -DDYNAMIC_PLOTTING -o es.o .././pari-2.1.7/src/language/es.cgcc: unrecognized option `-KPIC'gcc: language ildoff not recognizedgcc: .././pari-2.1.7/src/language/es.c: linker input file unused because linking not donegcc -c -I .././pari-2.1.7/src -I .././pari-2.1.7/src/headers -I .././pari-2.1.7/src/graph -I . -D_LARGEFILE_SOURCE -D_FILE_OFFSET_BITS=64 -xarch=v8 -D_TS_ERRNO -xO3 -xspace -xildoff -DVERSION=\"\" -DXS_VERSION=\"\" -KPIC -Derr=pari_err -DSOLARIS -DHAS_STAT -DHAS_OPENDIR -DDYNAMIC_PLOTTING -o base5.o .././pari-2.1.7
> /src/basemath/base5.cgcc: unrecognized option `-KPIC'gcc: language ildoff not recognizedgcc: .././pari-2.1.7/src/basemath/base5.c: linker input file unused because linking not donegcc -c -I .././pari-2.1.7/src -I .././pari-2.1.7/src/headers -I .././pari-2.1.7/src/graph -I . -D_LARGEFILE_SOURCE -D_FILE_OFFSET_BITS=64 -xarch=v8 -D_TS_ERRNO -xO3 -xspace -xildoff -DVERSION=\"\" -DXS_VERSION=\"\" -KPIC -Derr=pari_err -DSOLARIS -DHAS_STAT -DHAS_OPENDIR -DDYNAMIC_PLOTTING -o bibli1.o .././pari-2.1.7/src/basemath/bibli1.cgcc: unrecognized option `-KPIC'gcc: language ildoff not recognizedgcc: .././pari-2.1.7/src/basemath/bibli1.c: linker input file unused because linking not donegcc -c -I .././pari-2.1.7/src -I .././pari-2.1.7/src/headers -I .././pari-2.1.7/src/graph -I . -D_LARGEFILE_SOURCE -D_FILE_OFFSET_BITS=64 -xarch=v8 -D_TS_ERRNO -xO3 -xspace -xildoff -DVERSION=\"\" -DXS_VERSION=\"\" -KPIC -Derr=pari_err -DSOLARIS -DHAS_STAT -DHAS_OPENDIR -DDYNAMIC_PLOTTING -o buch2
a) It would make more sense if your copy-and-paste were less garbled.
b) In my new release (Math-Pari-2.010809b.tar.gz) I put some
workaround about configuration bugs in Solaris’ vendor’s Perl
(following the discussion in
http://prefetch.net/blog/index.php/2006/07/29/building-perl-modules-for-solaris/).
c) These workaround have a chance to fix your problems.
Could you please retest? (My release is already indexed, and
should propagate to CPAN in about a day.)
Thanks,
Ilya