W
W. Van Hooste
problem
I try to install Net::SSH:erl (instead of Net::SSH). And have lots
of problems with the compilation of prereq Math:ari and GP/PARI...
environment
I'm installing on a RS/6000 with a PPC604 processor but want to do the
same on more recent ones like POWER4 (I am using AIX 5.1).
details
No idea if the problem is GP/PARI or Math:ari but GP/PARI returns
some incorrect values but installes, and Math:ari failes during the
"make test"...
The AIX detection in GP/PARI does not seem to be correct. I tried
editing arch-osname and the Makefile and even tried the GP/PARI alpha
version - withouth improved results.
==> I think this should be "RS/6000" or "Power" instead of "aix"
....Processor of family `aix' detected
#### Do not know how to build for assembler `aix'. ####
#### Reversing to assembler-less type `port'. ####
#### ####
#### If you think your processor's assembler is supported ####
#### by PARI, edit libPARI/Makefile.PL and report. ####
#### ####
#### Alternatively, specify machine=YOURTYPE on the ####
#### perl Makefile.PL line ####
#### Recognized types: ####
#### alpha hppa m86k none sparcv7 sparcv8 sparcv8_micro ####
#### sparcv8_super ix86 ####
==>During compilation of Math:ari I notice that it completely
ignores the fact i have no native C-compiler but use gcc...
cc: unrecognized option `-qmaxmem=16384'
cc: unrecognized option `-q32'
cc: unrecognized option `-qlonglong'
==>but it compiles nevertheless!
Target "all" is up to date.
....
Target "makemakerdflt" is up to date.
/usr/bin/make -- OK
==>And then we have the dissapointment...
t/00_Pari...........ok 127/127FAILED test 109
Failed 1/127 tests, 99.21% okay
t/PlotRect..........# Can't locate Term/Gnuplot.pm in @INC, ignoring
the test
skipped: Can't locate Term/Gnuplot.pm in @INC
t/analyz............ok
t/elliptic..........ok
t/graph.............# Can't locate Term/Gnuplot.pm in @INC, ignoring
plotting
t/graph.............ok, 1/28 skipped: plot() - can't test it yet
t/linear............ok
t/nfields...........ok 23/161PARI: *** Warning: insufficient
precision for f
t/nfields...........ok 156/161PARI: *** Warning: not a fundamental
discrimin
t/nfields...........ok
t/number............ok
t/objets............ok, 1/62 skipped: test producing error unsupported
yet (impo
t/ploth.............# Can't locate Term/Gnuplot.pm in @INC, ignoring
plotting
t/ploth.............ok, 10/34 skipped: various reasons
t/polyser...........ok
t/program...........ok, 13/37 skipped: various reasons
t/sumiter...........ok
t/trans.............ok
Failed Test Status Wstat Total Fail Failed List of failed
-------------------------------------------------------------------------------
t/00_Pari.t 127 1 0.79% 109
1 test and 25 subtests skipped.
Failed 1/14 test scripts, 92.86% okay. 1/823 subtests failed, 99.88%
okay.
make: The error code from the last command is 255.
Stop.
/usr/bin/make test -- NOT OK
Running make install
make test had returned bad status, won't install without force
==>I did try force install and then installation of Crypt::Random
(prereq) failed with PARI errors!
All help is welcome!
I try to install Net::SSH:erl (instead of Net::SSH). And have lots
of problems with the compilation of prereq Math:ari and GP/PARI...
environment
I'm installing on a RS/6000 with a PPC604 processor but want to do the
same on more recent ones like POWER4 (I am using AIX 5.1).
details
No idea if the problem is GP/PARI or Math:ari but GP/PARI returns
some incorrect values but installes, and Math:ari failes during the
"make test"...
The AIX detection in GP/PARI does not seem to be correct. I tried
editing arch-osname and the Makefile and even tried the GP/PARI alpha
version - withouth improved results.
==> I think this should be "RS/6000" or "Power" instead of "aix"
....Processor of family `aix' detected
#### Do not know how to build for assembler `aix'. ####
#### Reversing to assembler-less type `port'. ####
#### ####
#### If you think your processor's assembler is supported ####
#### by PARI, edit libPARI/Makefile.PL and report. ####
#### ####
#### Alternatively, specify machine=YOURTYPE on the ####
#### perl Makefile.PL line ####
#### Recognized types: ####
#### alpha hppa m86k none sparcv7 sparcv8 sparcv8_micro ####
#### sparcv8_super ix86 ####
==>During compilation of Math:ari I notice that it completely
ignores the fact i have no native C-compiler but use gcc...
cc: unrecognized option `-qmaxmem=16384'
cc: unrecognized option `-q32'
cc: unrecognized option `-qlonglong'
==>but it compiles nevertheless!
Target "all" is up to date.
....
Target "makemakerdflt" is up to date.
/usr/bin/make -- OK
==>And then we have the dissapointment...
t/00_Pari...........ok 127/127FAILED test 109
Failed 1/127 tests, 99.21% okay
t/PlotRect..........# Can't locate Term/Gnuplot.pm in @INC, ignoring
the test
skipped: Can't locate Term/Gnuplot.pm in @INC
t/analyz............ok
t/elliptic..........ok
t/graph.............# Can't locate Term/Gnuplot.pm in @INC, ignoring
plotting
t/graph.............ok, 1/28 skipped: plot() - can't test it yet
t/linear............ok
t/nfields...........ok 23/161PARI: *** Warning: insufficient
precision for f
t/nfields...........ok 156/161PARI: *** Warning: not a fundamental
discrimin
t/nfields...........ok
t/number............ok
t/objets............ok, 1/62 skipped: test producing error unsupported
yet (impo
t/ploth.............# Can't locate Term/Gnuplot.pm in @INC, ignoring
plotting
t/ploth.............ok, 10/34 skipped: various reasons
t/polyser...........ok
t/program...........ok, 13/37 skipped: various reasons
t/sumiter...........ok
t/trans.............ok
Failed Test Status Wstat Total Fail Failed List of failed
-------------------------------------------------------------------------------
t/00_Pari.t 127 1 0.79% 109
1 test and 25 subtests skipped.
Failed 1/14 test scripts, 92.86% okay. 1/823 subtests failed, 99.88%
okay.
make: The error code from the last command is 255.
Stop.
/usr/bin/make test -- NOT OK
Running make install
make test had returned bad status, won't install without force
==>I did try force install and then installation of Crypt::Random
(prereq) failed with PARI errors!
All help is welcome!