This is the mail archive of the gsl-discuss@sources.redhat.com mailing list for the GSL project.


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]
Other format: [Raw text]

Re: make check fails with gcc 2.95


-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

On Friday 21 November 2003 12:57, Brian Gough wrote:
> Ivo Alxneit writes:
>  > hi everybody i ran `make check` (gsl1.4) on a rh7.3 system BUT WITH
>  > GCC DOWNGRADED TO 2.95
>
> Hi,
>
> Thanks for the bug report.
>
> Is the downgraded gcc officially supplied by Redhat, or did it come
> from somewhere else?

well semi-official i guess. at least its in the /contrib section on mirrors.

btw i redid the test both with an .i686 and an .i386 rpm and both fail in the 
same way. i guess this is one more reason to update to 9.x when i have time.
>
> I have not seen those errors before.  The version of the compiler
> should be ok.  I use gcc version 2.95.4 20011002 (Debian prerelease)
>
> I can only suggest using the debugger to take a closer look and see
> whether the behavior is valid or not, as you step through the code.
> For example to see where the value 5.80888e+17 in the first error line
> comes from, because it looks like a corrupt value.

probably no time for deeper investigations :-( but i will let you know

>
>  > finally the tests in /eigen (herm(x)...) pass but
>  >
>  > FAIL: test gsl_ran_tdist1 [-1,-0.9) (1940/100000=0.0194 observed vs
>  > 5.80888e+17 expected)

isn't this error message misleading? i would think 1940/100000=0.0194 is the 
correct number (i.e. expected) and 5.80888e+17 is wrong (thus observed in my 
test). 

>  >
>  > p.s. how can i continue with `make check` if i get errors in `make
>  > check`? for the above i manually removed all references to randist
>  > (lines/words) and fft.  now tests in cdf fail because i previously
>  > removed the reference to librandist.la.
>
> make -k check will continue after errors
> use make -k check >log 2>&1
thanks (RTFM istn't it? :-) )
- -- 
Dr. Ivo Alxneit
Laboratory for Solar Technology   phone: +41 56 310 4092
Paul Scherrer Institute             fax: +41 56 310 2624
CH-5232 Villigen                   http://solar.web.psi.ch
Switzerland                   gnupg key: 0x515E30C7
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.2.3 (GNU/Linux)

iD8DBQE/wbxxAd7CE1FeMMcRAuFqAKC9owj8MoiHQMOJJV6GEdHdfNsMkgCeLBRo
5UapJt8yzYQIA8kQPnzlqMk=
=RNiN
-----END PGP SIGNATURE-----


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]