Re: [Bug-gsl] GSL: make check fails

[Puen-Lap . Lee]

Brian,
        I changed the option to "-g -o2" and "-o2" both failed in the same 
way.
        Puenlap Lee.


make  test

make[2]: Entering directory 
`/homes/tmps/leepl/SciStudio/rpmbuild/BUILD/ia64/GSL-1.12/gsl-1.12/statistics'

gcc -DHAVE_CONFIG_H -I. -I.. -I..    -g -O2 -c test.c
gcc -DHAVE_CONFIG_H -I. -I.. -I..    -g -O2 -c test_nist.c

/bin/sh ../libtool --tag=CC   --mode=link gcc  -g -O2   -o test test.o 
test_nist.o libgslstatistics.la ../sort/libgslsort.la 
../ieee-utils/libgslieeeutils.la ..
/err/libgslerr.la ../test/libgsltest.la ../sys/libgslsys.la 
../utils/libutils.la -lm

libtool: link: gcc -g -O2 -o test test.o test_nist.o 
./.libs/libgslstatistics.a ../sort/.libs/libgslsort.a 
../ieee-utils/.libs/libgslieeeutils.a ../err/.libs/l
ibgslerr.a ../test/.libs/libgsltest.a ../sys/.libs/libgslsys.a 
../utils/.libs/libutils.a -lm

make[2]: Leaving directory 
`/homes/tmps/leepl/SciStudio/rpmbuild/BUILD/ia64/GSL-1.12/gsl-1.12/statistics'

make  check-TESTS

make[2]: Entering directory 
`/homes/tmps/leepl/SciStudio/rpmbuild/BUILD/ia64/GSL-1.12/gsl-1.12/statistics'

FAIL: gsl_stats_long_double_max NaN (nan observed vs nan expected) [128]
FAIL: gsl_stats_long_double_min NaN (nan observed vs nan expected) [129]
FAIL: gsl_stats_long_double_max NaN (nan observed vs nan expected) [479]
FAIL: gsl_stats_long_double_min NaN (nan observed vs nan expected) [480]
FAIL: gsl_stats_long_double_max NaN (nan observed vs nan expected) [830]
FAIL: gsl_stats_long_double_min NaN (nan observed vs nan expected) [831]
FAIL: test

===================

1 of 1 tests failed

===================

make[2]: *** [check-TESTS] Error 1



Brian Gough <address@hidden> 
08/13/2009 05:15 AM
Please respond to
address@hidden


To
address@hidden
cc
address@hidden
Subject
Re: [Bug-gsl] GSL: make check fails






At Thu, 6 Aug 2009 15:44:42 -0700,
address@hidden wrote:
> I ran a "./configure --prefix=/my-install-folder" with no error
> I did a 'export CFLAGS="-O3"' then the make with no error, but the "make 

> check" report error in the during "Making check in statistics". 
> 
> FAIL: gsl_stats_long_double_max NaN (nan observed vs nan expected) [128]
> FAIL: gsl_stats_long_double_min NaN (nan observed vs nan expected) [129]
> FAIL: gsl_stats_long_double_max NaN (nan observed vs nan expected) [479]
> FAIL: gsl_stats_long_double_min NaN (nan observed vs nan expected) [480]
> FAIL: gsl_stats_long_double_max NaN (nan observed vs nan expected) [830]
> FAIL: gsl_stats_long_double_min NaN (nan observed vs nan expected) [831]
> FAIL: test
> 

Thanks for the bug report.  Does the same problem occur if you compile
without optimisation? (or with the default optimisation of -O2 instead
of -O3).  From the output it appears that the result is correct, but
the isnan() call used to test it is failing.

-- 
Brian Gough
(GSL Maintainer)

Support freedom by joining the FSF 
http://www.fsf.org/associate/support_freedom/join_fsf?referrer=37