new testing version 1.7.90

new testing version 1.7.90

[Brian Gough]

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

I'm preparing a new release... please try it out and post back 
what platforms it works or doesn't work on. Thanks.

http://network-theory.co.uk/download/gsl/gsl-1.7.90.tar.gz
http://network-theory.co.uk/download/gsl/gsl-1.7.90.tar.gz.sig

$ md5sum gsl-1.7.90*
deffa8607840c822a906eb89ad0323f1  gsl-1.7.90.tar.gz
a986ddbcbac26a6e520ae89aa91dd611  gsl-1.7.90.tar.gz.sig

-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.0 (GNU/Linux)

iD8DBQFEIrQ4biFv7WQGnVwRAopmAJwKE+YfcXJgzVVtaHeDAJ07GR5voQCeIyq4
7sUyhYI8l/cN3rHozi4LF8I=
=esPF
-----END PGP SIGNATURE-----

Re: new testing version 1.7.90

[David Geldreich]

Hi all,

Brian Gough a écrit :
> -----BEGIN PGP SIGNED MESSAGE-----
> Hash: SHA1
> 
> I'm preparing a new release... please try it out and post back 
> what platforms it works or doesn't work on. Thanks.

I have made a port of this testing version of GSL 1.7.90 for Windows 
Visual Studio .NET 2003 :

http://david.geldreich.free.fr/dev.html
http://david.geldreich.free.fr/downloads/gsl-1.7.90-windows-sources.zip
http://david.geldreich.free.fr/downloads/gsl-1.7.90-windows-binaries.zip

tests failed for the test_statistics with messages like :

FAIL: gsl_stats_long_double_max NaN (0.1331 observed vs -1.#IND 
expected) [113]
FAIL: gsl_stats_long_double_min NaN (0.0243 observed vs -1.#IND 
expected) [114]
FAIL: gsl_stats_long_double_minmax max NaN (0.1331 observed vs -1.#IND 
expected) [115]
FAIL: gsl_stats_long_double_minmax min NaN (0.0243 observed vs -1.#IND 
expected) [116]
FAIL: gsl_stats_long_double_max_index NaN (4 observed vs 3 expected) [117]
FAIL: gsl_stats_long_double_min_index NaN (6 observed vs 3 expected) [118]
FAIL: gsl_stats_long_double_minmax_index max NaN (4 observed vs 3 
expected) [119]
FAIL: gsl_stats_long_double_minmax_index min NaN (6 observed vs 3 
expected) [120]

I will try to find why.

Thanks for your hard work on GSL.

Re: new testing version 1.7.90

[David Geldreich]

David Geldreich a écrit :
> I have made a port of this testing version of GSL 1.7.90 for Windows 
> Visual Studio .NET 2003 :
> 
> http://david.geldreich.free.fr/dev.html
> http://david.geldreich.free.fr/downloads/gsl-1.7.90-windows-sources.zip
> http://david.geldreich.free.fr/downloads/gsl-1.7.90-windows-binaries.zip
> 
> tests failed for the test_statistics with messages like :
> 
> FAIL: gsl_stats_long_double_max NaN (0.1331 observed vs -1.#IND 
> expected) [113]
> FAIL: gsl_stats_long_double_min NaN (0.0243 observed vs -1.#IND 
> expected) [114]

I have corrected this problem (my cvs import erased the #define FP 1 in 
template_on.h). This allows gsl_stats minmax to deal with NaN numbers.

But I also got :

FAIL: gsl_sf_bessel_In_scaled_e( 2, 1e8, &r) [117]
   expected: 3.989422729213446e-005
   obtained: 3.98942272921265e-005   6.233473006603277e-007  0.0153846
   fracdiff: 9.979024842224512e-014
   value not within tolerance of expected value
    3.98942272921265e-005  6.233473006603277e-007
FAIL: Bessel Functions [394]

Except this one, all the tests are passing OK under Windows Visual .NET 2003

Re: new testing version 1.7.90

[Jochen Küpper]

[-- Attachment #1: Type: text/plain, Size: 902 bytes --]

Brian Gough <bjg@network-theory.co.uk> writes:

> I'm preparing a new release... please try it out and post back 
> what platforms it works or doesn't work on. Thanks.

I've tried cvs-HEAD on several Linux platforms (Pentium4 and AMD
Opteron) running Fedora Core 3/4, SuSe Enterprise Server 9, using
gcc-3.3, gcc-4.0, and gcc-4.1. 

make && make check runs successfully on all of them.

As a side-remark: On the Opterons using gcc-4 I get many of the
following warning messages:
,----
| ../../../source/gsl/linalg/test.c:489: warning: \
| format '%3d' expects type 'int', but argument 2 has type 'size_t'
`----

Thank you for continued work on GSL!

Greetings,
Jochen
-- 
Einigkeit und Recht und Freiheit                http://www.Jochen-Kuepper.de
    Liberté, Égalité, Fraternité                GnuPG key: CC1B0B4D
        (Part 3 you find in my messages before fall 2003.)

[-- Attachment #2: Type: application/pgp-signature, Size: 188 bytes --]

Re: new testing version 1.7.90

[Stefan Gerlach]

[-- Attachment #1: Type: text/plain, Size: 2184 bytes --]

Hi,

I test 1.7.90 on serveral Linux distributions using CFLAGS="-O3 -Wall" and 
"make" & "make check" shows no errors. But i got some warnings:

1. On gcc-4 systems (FC4+5, SUSE 10.0, Mandriva 2006) :
$ make
...
coulomb.c: In function 'gsl_sf_coulomb_wave_FG_e':
coulomb.c:994: warning: 'G_lam_G' is used uninitialized in this function
coulomb.c:996: warning: 'Gp_lam_G' is used uninitialized in this function
...
fermi_dirac.c: In function 'fd_neg':
fermi_dirac.c:946: warning: 's' may be used uninitialized in this function
...

$ make check
...
make[2]: Entering directory `/sw/gsl-1.7.90/diff'
gcc -DHAVE_CONFIG_H -I. -I. -I.. -I..    -O3 -Wall -c test.c
test.c: In function 'main': 
test.c:137: warning: 'DF1.params' is used uninitialized in this function
test.c:137: warning: 'DF2.params' is used uninitialized in this function
test.c:137: warning: 'DF3.params' is used uninitialized in this function
test.c:137: warning: 'DF4.params' is used uninitialized in this function
test.c:137: warning: 'DF5.params' is used uninitialized in this function
test.c:137: warning: 'DF6.params' is used uninitialized in this function
...

2. during "configure" on SUSE 10.0 and SLES 9 i get :
...
checking whether the gcc linker (/usr/i586-suse-linux/bin/ld) supports shared libraries... yes
checking whether -lc should be explicitly linked in... no
checking dynamic linker characteristics... cat: /etc/ld.so.conf.d/*.conf: No such file or directory
GNU/Linux ld.so
...

But that may be a bug of SUSE.

3. When compiling with "make -j 2" (testet on SUSE 9.1 dual P4) i get :
...
Making all in permutation
make[2]: Entering directory `/tmp/gsl-1.7.90/permutation'
/bin/sh ../libtool --tag=CC --mode=compile gcc -DHAVE_CONFIG_H -I. -I. -I.. -I.. -I..    -O3 -Wall -c -o init.lo init.c
/bin/sh ../libtool --tag=CC --mode=compile gcc -DHAVE_CONFIG_H -I. -I. -I.. -I.. -I..    -O3 -Wall -c -o file.lo file.c
mkdir .libs
mkdir .libs
mkdir: cannot create directory `.libs': File exists
...

The same for the directories linalg/, fft/, min/, monte/ and permutation/.


Hope this helps in any way.

Thanks for your work.

-- 
Stefan Gerlach

[-- Attachment #2: Type: application/pgp-signature, Size: 189 bytes --]

Re: new testing version 1.7.90

[Jason Stover]

> uname -a
NetBSD ukato 2.1.0_STABLE NetBSD 2.1.0_STABLE (sdf) #0: Tue Mar 14 07:50:31 UTC 2006  root@sverige:/var/src/src/sys/arch/alpha/compile/sdf alpha

The following test failed:

Making check in ieee-utils
/usr/bin/make  test
gcc -DHAVE_CONFIG_H -I. -I. -I.. -I..     -g -O2 -c test.c
/usr/pkg/bin/bash ../libtool --tag=CC --mode=link gcc  -g -O2    -o test  test.o libgslieeeutils.la ../err/libgslerr.la ../test/libgsltest.la ../sys/libgslsys.la ../utils/libutils.la -lm
gcc -g -O2 -o test test.o  ./.libs/libgslieeeutils.a ../err/.libs/libgslerr.a ../test/.libs/libgsltest.a ../sys/.libs/libgslsys.a ../utils/.libs/libutils.a -lm
/usr/bin/make  check-TESTS
FAIL: float x = -1*0, sign is - (0 observed vs 1 expected) [5]
[1]   Floating point exception (core dumped) ${dir}${tst}
FAIL: test

-Jason

On Thu, Mar 23, 2006 at 02:44:29PM +0000, Brian Gough wrote:
> -----BEGIN PGP SIGNED MESSAGE-----
> Hash: SHA1
> 
> I'm preparing a new release... please try it out and post back 
> what platforms it works or doesn't work on. Thanks.
> 
> http://network-theory.co.uk/download/gsl/gsl-1.7.90.tar.gz
> http://network-theory.co.uk/download/gsl/gsl-1.7.90.tar.gz.sig
> 
> $ md5sum gsl-1.7.90*
> deffa8607840c822a906eb89ad0323f1  gsl-1.7.90.tar.gz
> a986ddbcbac26a6e520ae89aa91dd611  gsl-1.7.90.tar.gz.sig
> 
> -----BEGIN PGP SIGNATURE-----
> Version: GnuPG v1.4.0 (GNU/Linux)
> 
> iD8DBQFEIrQ4biFv7WQGnVwRAopmAJwKE+YfcXJgzVVtaHeDAJ07GR5voQCeIyq4
> 7sUyhYI8l/cN3rHozi4LF8I=
> =esPF
> -----END PGP SIGNATURE-----

-- 
jstover@sdf.lonestar.org
SDF Public Access UNIX System - http://sdf.lonestar.org

Re: new testing version 1.7.90

[Jason Stover]

> uname -ap
OpenBSD math.gcsu.edu 3.8 GENERIC#138 i386 Intel(R) Xeon(TM) CPU 2.80GHz ("GenuineIntel" 686-class)

All tests passed.

On Thu, Mar 23, 2006 at 02:44:29PM +0000, Brian Gough wrote:
> -----BEGIN PGP SIGNED MESSAGE-----
> Hash: SHA1
> 
> I'm preparing a new release... please try it out and post back 
> what platforms it works or doesn't work on. Thanks.
> 
> http://network-theory.co.uk/download/gsl/gsl-1.7.90.tar.gz
> http://network-theory.co.uk/download/gsl/gsl-1.7.90.tar.gz.sig
> 
> $ md5sum gsl-1.7.90*
> deffa8607840c822a906eb89ad0323f1  gsl-1.7.90.tar.gz
> a986ddbcbac26a6e520ae89aa91dd611  gsl-1.7.90.tar.gz.sig
> 
> -----BEGIN PGP SIGNATURE-----
> Version: GnuPG v1.4.0 (GNU/Linux)
> 
> iD8DBQFEIrQ4biFv7WQGnVwRAopmAJwKE+YfcXJgzVVtaHeDAJ07GR5voQCeIyq4
> 7sUyhYI8l/cN3rHozi4LF8I=
> =esPF
> -----END PGP SIGNATURE-----

-- 
jstover@sdf.lonestar.org
SDF Public Access UNIX System - http://sdf.lonestar.org

Re: new testing version 1.7.90

[Stefan Gerlach]

[-- Attachment #1: Type: text/plain, Size: 2919 bytes --]

Hi Again,

I did some further testing and found some problems:

1. Solaris 8, SPARC, gcc 3.3.1 and Solaris 10, x86, gcc 3.4.2:
...
Making all in siman
gmake[2]: Entering directory `/tmp/gsl-1.7.90/siman'
/bin/bash ../libtool --tag=CC --mode=link gcc  -O3 -Wall   -o siman_tsp
siman_tsp.o libgslsiman.la ../rng/libgslrng.la
../ieee-utils/libgslieeeutils.la ../err/libgslerr.la ../utils/libutils.la -lm 
gcc -O3 -Wall -o siman_tsp siman_tsp.o  ./.libs/libgslsiman.a
../rng/.libs/libgslrng.a ../ieee-utils/.libs/libgslieeeutils.a
../err/.libs/libgslerr.a ../utils/.libs/libutils.a -lm
gcc: ./.libs/libgslsiman.a: No such file or directory
gcc: ../rng/.libs/libgslrng.a: No such file or directory
gcc: ../ieee-utils/.libs/libgslieeeutils.a: No such file or directory
gcc: ../err/.libs/libgslerr.a: No such file or directory
gcc: ../utils/.libs/libutils.a: No such file or directory
gmake[2]: *** [siman_tsp] Error 1 
...

The same on Solaris 8, SPARC, cc:
...
ild: (bad file) argument file doesn't exist :: ./.libs/libgslsiman.a
gmake[2]: *** [siman_tsp] Error 1 
...

2. SUSE 10.0/9.1, x86, icc 9.0 :

...
make  check-TESTS
make[2]: Entering directory `/tmp/gsl-1.7.90/statistics'
FAIL: gsl_stats_min NaN (0.0243 observed vs nan expected) [34]
FAIL: gsl_stats_float_min NaN (0.0243 observed vs nan expected) [74]
FAIL: gsl_stats_long_double_min NaN (0.0243 observed vs nan expected) [114]
FAIL: gsl_stats_min NaN (0.0243 observed vs nan expected) [362]
FAIL: gsl_stats_float_min NaN (0.0243 observed vs nan expected) [402]
FAIL: gsl_stats_long_double_min NaN (0.0243 observed vs nan expected) [442]
FAIL: gsl_stats_min NaN (0.0243 observed vs nan expected) [690]
FAIL: gsl_stats_float_min NaN (0.0243 observed vs nan expected) [730]
FAIL: gsl_stats_long_double_min NaN (0.0243 observed vs nan expected) [770]
...

3. RedHat 7.1 ALPHA EV6, gcc/ccc and Debian 2.2, ALPHA EV67, gcc :

$ make
...
jacobi.c: In function `gsl_eigen_invert_jacobi':
jacobi.c:227: warning: passing arg 5 of `gsl_eigen_jacobi' from incompatible 
pointer type
...

$ make check		(also mentioned by Jason)
...
make  check-TESTS
make[2]: Entering directory `/tmp/gsl-1.7.90/ieee-utils'
FAIL: float x = -1*0, sign is - (0 observed vs 1 expected) [5]
FAIL: test
...

But many other tests just segfault!
for instance in integration/

...
$ gdb ./test core
#0  0x120015e9c in compute_moments (par=8.4785386524524643e-78, 
chebmo=0x2000024eb30) at qmomof.c:174
174       v[2] = (32 * (par2 - 12) * cospar
(gdb) bt
#0  0x120015e9c in compute_moments (par=8.4785386524524643e-78, 
chebmo=0x2000024eb30) at qmomof.c:174
#1  0x120015a2c in gsl_integration_qawo_table_alloc (omega=31.415926535897931, 
L=1, sine=GSL_INTEG_SINE, n=1000) at qmomof.c:78
#2  0x120008fdc in main () at test.c:1924
...

BTW: I had no problems on all other platforms i test it :-)

-- 
Stefan Gerlach


[-- Attachment #2: Type: application/pgp-signature, Size: 189 bytes --]

Re: new testing version 1.7.90

[Brian Gough]

Stefan Gerlach writes:
 > 1. Solaris 8, SPARC, gcc 3.3.1 and Solaris 10, x86, gcc 3.4.2:
 > ...
 > Making all in siman
 > gmake[2]: Entering directory `/tmp/gsl-1.7.90/siman'
 > /bin/bash ../libtool --tag=CC --mode=link gcc  -O3 -Wall   -o siman_tsp
 > siman_tsp.o libgslsiman.la ../rng/libgslrng.la
 > ../ieee-utils/libgslieeeutils.la ../err/libgslerr.la ../utils/libutils.la -lm 
 > gcc -O3 -Wall -o siman_tsp siman_tsp.o  ./.libs/libgslsiman.a
 > ../rng/.libs/libgslrng.a ../ieee-utils/.libs/libgslieeeutils.a
 > ../err/.libs/libgslerr.a ../utils/.libs/libutils.a -lm
 > gcc: ./.libs/libgslsiman.a: No such file or directory
 > gcc: ../rng/.libs/libgslrng.a: No such file or directory
 > gcc: ../ieee-utils/.libs/libgslieeeutils.a: No such file or directory
 > gcc: ../err/.libs/libgslerr.a: No such file or directory
 > gcc: ../utils/.libs/libutils.a: No such file or directory
 > gmake[2]: *** [siman_tsp] Error 1 

Looks like it's not making any libraries at all ... can you check the
section "Solaris Hints" in the install file.  Sometimes this is
because it did not find 'ar' during configure or something like
that. Thanks.

-- 
Brian Gough

Re: new testing version 1.7.90

[Stefan Gerlach]

[-- Attachment #1: Type: text/plain, Size: 1435 bytes --]

After defining RANLIB=granlib, AR=gar and MAKE=gmake "gmake" and "gmake check" 
works now.
Shouldn't "configure" print out an error message if "ar" wasn't found?

-- 
Stefan Gerlach

On Monday 27 March 2006 12:43, you wrote:
> Stefan Gerlach writes:
>  > 1. Solaris 8, SPARC, gcc 3.3.1 and Solaris 10, x86, gcc 3.4.2:
>  > ...
>  > Making all in siman
>  > gmake[2]: Entering directory `/tmp/gsl-1.7.90/siman'
>  > /bin/bash ../libtool --tag=CC --mode=link gcc  -O3 -Wall   -o siman_tsp
>  > siman_tsp.o libgslsiman.la ../rng/libgslrng.la
>  > ../ieee-utils/libgslieeeutils.la ../err/libgslerr.la
>  > ../utils/libutils.la -lm gcc -O3 -Wall -o siman_tsp siman_tsp.o 
>  > ./.libs/libgslsiman.a
>  > ../rng/.libs/libgslrng.a ../ieee-utils/.libs/libgslieeeutils.a
>  > ../err/.libs/libgslerr.a ../utils/.libs/libutils.a -lm
>  > gcc: ./.libs/libgslsiman.a: No such file or directory
>  > gcc: ../rng/.libs/libgslrng.a: No such file or directory
>  > gcc: ../ieee-utils/.libs/libgslieeeutils.a: No such file or directory
>  > gcc: ../err/.libs/libgslerr.a: No such file or directory
>  > gcc: ../utils/.libs/libutils.a: No such file or directory
>  > gmake[2]: *** [siman_tsp] Error 1
>
> Looks like it's not making any libraries at all ... can you check the
> section "Solaris Hints" in the install file.  Sometimes this is
> because it did not find 'ar' during configure or something like
> that. Thanks.

[-- Attachment #2: Type: application/pgp-signature, Size: 189 bytes --]

Re: new testing version 1.7.90

[Brian Gough]

Stefan Gerlach writes:
 > After defining RANLIB=granlib, AR=gar and MAKE=gmake "gmake" and "gmake check" 
 > works now.
 > Shouldn't "configure" print out an error message if "ar" wasn't found?

Can you try this update and tell me if it makes any difference:

http://network-theory.co.uk/download/gsl/gsl-1.7.91.tar.gz
http://network-theory.co.uk/download/gsl/gsl-1.7.91.tar.gz.sig

If it still happens I'd like to see the complete configure output,
config.log and the PATH.

I've looked into the other errors on SUSE, they are from libtool but
are just stray messages, they don't affect the compilation.

checking dynamic linker characteristics... cat: /etc/ld.so.conf.d/*.conf: No such file or directory

mkdir .libs
mkdir: cannot create directory `.libs': File exists

-- 
Brian Gough

Re: new testing version 1.7.90

[Stefan Gerlach]

[-- Attachment #1.1: Type: text/plain, Size: 2108 bytes --]

Sorry, but it doesn't work as expected.

$ ./configure
...
checking for ar... false
checking for ranlib... :
checking for strip... :
...

$ gmake
...
gmake[2]: Entering directory `/tmp/gsl-1.7.91/utils'
/bin/bash ../libtool --tag=CC --mode=compile gcc -DHAVE_CONFIG_H -I. -I. -I..     -g -O2 -c -o placeholder.lo placeholder.c
mkdir .libs
 gcc -DHAVE_CONFIG_H -I. -I. -I.. -g -O2 -c placeholder.c  -fPIC -DPIC -o .libs/placeholder.o
 gcc -DHAVE_CONFIG_H -I. -I. -I.. -g -O2 -c placeholder.c -o placeholder.o >/dev/null 2>&1
/bin/bash ../libtool --tag=CC --mode=link gcc  -g -O2   -o libutils.la   placeholder.lo  -lm 
false cru .libs/libutils.a .libs/placeholder.o
gmake[2]: *** [libutils.la] Error 1
gmake[2]: Leaving directory `/tmp/gsl-1.7.91/utils'
gmake[1]: *** [all-recursive] Error 1
...

(the complete logs are attached)

Obviously ar and ranlib are necessary to compile gsl and configure should complain about it 
the same way it does for a missing gcc :

$ export CC=ccc && ./configure
...
checking for gcc... ccc
checking for C compiler default output file name... configure: error: C compiler cannot create executables
See `config.log' for more details.
$

-- 
Stefan Gerlach

On Wednesday 29 March 2006 14:57, Brian Gough wrote:
> Stefan Gerlach writes:
>  > After defining RANLIB=granlib, AR=gar and MAKE=gmake "gmake" and "gmake
>  > check" works now.
>  > Shouldn't "configure" print out an error message if "ar" wasn't found?
>
> Can you try this update and tell me if it makes any difference:
>
> http://network-theory.co.uk/download/gsl/gsl-1.7.91.tar.gz
> http://network-theory.co.uk/download/gsl/gsl-1.7.91.tar.gz.sig
>
> If it still happens I'd like to see the complete configure output,
> config.log and the PATH.
>
> I've looked into the other errors on SUSE, they are from libtool but
> are just stray messages, they don't affect the compilation.
>
> checking dynamic linker characteristics... cat: /etc/ld.so.conf.d/*.conf:
> No such file or directory
>
> mkdir .libs
> mkdir: cannot create directory `.libs': File exists

[-- Attachment #1.2: configure.log --]
[-- Type: text/x-log, Size: 6373 bytes --]

checking for a BSD-compatible install... /software/bin/install -c
checking whether build environment is sane... yes
checking for gawk... gawk
checking whether gmake sets $(MAKE)... yes
checking whether to enable maintainer-specific portions of Makefiles... no
checking whether gmake sets $(MAKE)... (cached) yes
checking build system type... sparc-sun-solaris2.8
checking host system type... sparc-sun-solaris2.8
checking for gcc... gcc
checking for C compiler default output file name... a.out
checking whether the C compiler works... yes
checking whether we are cross compiling... no
checking for suffix of executables... 
checking for suffix of object files... o
checking whether we are using the GNU C compiler... yes
checking whether gcc accepts -g... yes
checking for gcc option to accept ANSI C... none needed
checking how to run the C preprocessor... gcc -E
checking for a BSD-compatible install... /software/bin/install -c
checking whether ln -s works... yes
checking for a sed that does not truncate output... /opt/sfw/bin/gsed
checking for egrep... egrep
checking for ld used by gcc... /usr/ccs/bin/ld
checking if the linker (/usr/ccs/bin/ld) is GNU ld... no
checking for /usr/ccs/bin/ld option to reload object files... -r
checking for BSD-compatible nm... /usr/ccs/bin/nm -p
checking how to recognise dependent libraries... pass_all
checking for ANSI C header files... yes
checking for sys/types.h... yes
checking for sys/stat.h... yes
checking for stdlib.h... yes
checking for string.h... yes
checking for memory.h... yes
checking for strings.h... yes
checking for inttypes.h... yes
checking for stdint.h... no
checking for unistd.h... yes
checking dlfcn.h usability... yes
checking dlfcn.h presence... yes
checking for dlfcn.h... yes
checking the maximum length of command line arguments... 262144
checking command to parse /usr/ccs/bin/nm -p output from gcc object... ok
checking for objdir... .libs
checking for ar... false
checking for ranlib... :
checking for strip... :
checking if gcc supports -fno-rtti -fno-exceptions... no
checking for gcc option to produce PIC... -fPIC
checking if gcc PIC flag -fPIC works... yes
checking if gcc static flag -static works... yes
checking if gcc supports -c -o file.o... yes
checking whether the gcc linker (/usr/ccs/bin/ld) supports shared libraries... yes
checking whether -lc should be explicitly linked in... yes
checking dynamic linker characteristics... solaris2.8 ld.so
checking how to hardcode library paths into programs... immediate
checking whether stripping libraries is possible... no
checking if libtool supports shared libraries... yes
checking whether to build shared libraries... yes
checking whether to build static libraries... yes
configure: creating libtool
checking for size_t... yes
checking for working volatile... yes
checking for inline... inline
checking for extern inline... yes
checking ieeefp.h usability... yes
checking ieeefp.h presence... yes
checking for ieeefp.h... yes
checking for vprintf... yes
checking for _doprnt... yes
checking for memcpy... yes
checking for memmove... yes
checking for strdup... yes
checking for strtol... yes
checking for strtoul... yes
checking for EXIT_SUCCESS and EXIT_FAILURE... yes
checking for cos in -lm... yes
checking whether feenableexcept is declared... no
checking whether fesettrapenable is declared... no
checking whether hypot is declared... yes
checking whether expm1 is declared... yes
checking whether acosh is declared... yes
checking whether asinh is declared... yes
checking whether atanh is declared... yes
checking whether ldexp is declared... yes
checking whether frexp is declared... yes
checking whether isinf is declared... no
checking whether finite is declared... no
checking whether isfinite is declared... no
checking whether isnan is declared... yes
checking whether log1p is declared... yes
checking for long double stdio... yes
checking for extended floating point registers... no
checking for IEEE arithmetic interface type... solaris
checking for IEEE compiler flags... none
checking for IEEE comparisons... yes
checking for IEEE denormalized values... yes
configure: creating ./config.status
config.status: creating gsl-config
config.status: creating gsl.pc
config.status: creating gsl_version.h
config.status: creating gsl.spec
config.status: creating gsl/Makefile
config.status: creating test/Makefile
config.status: creating err/Makefile
config.status: creating sys/Makefile
config.status: creating utils/Makefile
config.status: creating const/Makefile
config.status: creating min/Makefile
config.status: creating multimin/Makefile
config.status: creating ieee-utils/Makefile
config.status: creating fft/Makefile
config.status: creating specfunc/Makefile
config.status: creating dht/Makefile
config.status: creating fit/Makefile
config.status: creating multifit/Makefile
config.status: creating statistics/Makefile
config.status: creating sum/Makefile
config.status: creating roots/Makefile
config.status: creating multiroots/Makefile
config.status: creating ntuple/Makefile
config.status: creating poly/Makefile
config.status: creating qrng/Makefile
config.status: creating rng/Makefile
config.status: creating randist/Makefile
config.status: creating siman/Makefile
config.status: creating integration/Makefile
config.status: creating interpolation/Makefile
config.status: creating doc/Makefile
config.status: creating block/Makefile
config.status: creating vector/Makefile
config.status: creating matrix/Makefile
config.status: creating histogram/Makefile
config.status: creating monte/Makefile
config.status: creating ode-initval/Makefile
config.status: creating cblas/Makefile
config.status: creating blas/Makefile
config.status: creating linalg/Makefile
config.status: creating eigen/Makefile
config.status: creating permutation/Makefile
config.status: creating combination/Makefile
config.status: creating sort/Makefile
config.status: creating complex/Makefile
config.status: creating diff/Makefile
config.status: creating deriv/Makefile
config.status: creating cheb/Makefile
config.status: creating cdf/Makefile
config.status: creating wavelet/Makefile
config.status: creating Makefile
config.status: creating config.h
config.status: config.h is unchanged

[-- Attachment #1.3: config.log.gz --]
[-- Type: application/x-gzip, Size: 7542 bytes --]

[-- Attachment #2: Type: application/pgp-signature, Size: 189 bytes --]

Re: new testing version 1.7.90

[Brian Gough]

Stefan Gerlach writes:
 > 3. RedHat 7.1 ALPHA EV6, gcc/ccc and Debian 2.2, ALPHA EV67, gcc :
 > ...
 > make  check-TESTS
 > make[2]: Entering directory `/tmp/gsl-1.7.90/ieee-utils'
 > FAIL: float x = -1*0, sign is - (0 observed vs 1 expected) [5]
 > FAIL: test

This seems to be a problem with Alpha.  I don't have access to any
alpha systems.  

Maybe somebody can figure it out... you might need some CFLAGS options
(see the Alpha section in the INSTALL file).

-- 
Brian Gough

Re: new testing version 1.7.90

[Stefan Gerlach]

[-- Attachment #1: Type: text/plain, Size: 1630 bytes --]

The compiler flag "-mieee" is needed to fix this problem with gcc and ccc on 
Alpha.

As the gcc manual says :
...
`-mieee'
     The Alpha architecture implements floating-point hardware
     optimized for maximum performance.  It is mostly compliant with
     the IEEE floating point standard.  However, for full compliance,
     software assistance is required.  This option generates code fully
     IEEE compliant code _except_ that the INEXACT-FLAG is not
     maintained (see below).  If this option is turned on, the
     preprocessor macro `_IEEE_FP' is defined during compilation.  The
     resulting code is less efficient but is able to correctly support
     denormalized numbers and exceptional IEEE values such as
     not-a-number and plus/minus infinity.  Other Alpha compilers call
     this option `-ieee_with_no_inexact'.
...

BTW: 
I think the file INSTALL isn't correct here.
On Alpha the Compaq C compiler is called "ccc", "cc" is just a link to "gcc". 
Only on Tru64 UNIX the Compaq C compiler is called "cc".

-- 
Stefan Gerlach

On Monday 27 March 2006 17:14, Brian Gough wrote:
> Stefan Gerlach writes:
>  > 3. RedHat 7.1 ALPHA EV6, gcc/ccc and Debian 2.2, ALPHA EV67, gcc :
>  > ...
>  > make  check-TESTS
>  > make[2]: Entering directory `/tmp/gsl-1.7.90/ieee-utils'
>  > FAIL: float x = -1*0, sign is - (0 observed vs 1 expected) [5]
>  > FAIL: test
>
> This seems to be a problem with Alpha.  I don't have access to any
> alpha systems.
>
> Maybe somebody can figure it out... you might need some CFLAGS options
> (see the Alpha section in the INSTALL file).

[-- Attachment #2: Type: application/pgp-signature, Size: 189 bytes --]

Re: new testing version 1.7.90

[John D Lamb]

Brian Gough wrote:
> I'm preparing a new release... please try it out and post back
> what platforms it works or doesn't work on. Thanks.

No problems (all tests passed) on Linux 2.6.11.4-21.11-default i686
GNU/Linux (SuSE 9.3) system, building in separate directory from source
with CFLAGS=-march=pentium4 -O2 -mfpmath=sse -msse -msse2 -malign-double
-Wall -pipe -pedantic -Wno-long-long (i.e. pentium 4 math optimised)

There are some 'mostly harmless' warnings if you used -ansi; the last
one I saw was for string lengths in gsl_randist.c.

-- 
JDL

Re: new testing version 1.7.90

[Giulio Bottazzi]

On Thu, 23 Mar 2006 14:44:29 +0000
Brian Gough <bjg@network-theory.co.uk> wrote:

> -----BEGIN PGP SIGNED MESSAGE-----
> Hash: SHA1
> 
> I'm preparing a new release... please try it out and post back 
> what platforms it works or doesn't work on. Thanks.

No problems (all test passed) on a Pentium M 1.50GHz and a dual Opteron
242 both with Linux and

kernel: 2.6.15-gentoo-r1 
glibc version: :1.2.10

G.

Re: new testing version 1.7.90

[Heiko Bauke]

Hi,

On Thu, 23 Mar 2006 14:44:29 +0000
Brian Gough <bjg@network-theory.co.uk> wrote:

> I'm preparing a new release... please try it out and post back 
> what platforms it works or doesn't work on. Thanks.

"make check" fails on gcc 3.3.5 with options "-ansi -pedantic" because
some test uses the non-standard macro M_PI.


Making check in const
make[1]: Gehe in Verzeichnis »/home/bauke/tmp/gsl-1.7.90/const«
make  test
make[2]: Gehe in Verzeichnis »/home/bauke/tmp/gsl-1.7.90/const«
gcc -DHAVE_CONFIG_H -I. -I. -I.. -I..    -O3 -ansi -pedantic -c test.c
test.c: In Funktion »main«:
test.c:87: error: `M_PI' undeclared (first use in this function)
test.c:87: error: (Each undeclared identifier is reported only once
test.c:87: error: for each function it appears in.)
make[2]: *** [test.o] Fehler 1
make[2]: Verlasse Verzeichnis »/home/bauke/tmp/gsl-1.7.90/const«
make[1]: *** [check-am] Fehler 2
make[1]: Verlasse Verzeichnis »/home/bauke/tmp/gsl-1.7.90/const«
make: *** [check-recursive] Fehler 1


	With regards

	Heiko Bauke


-- 
-- Universitäten sind Bildungsstätten, die aus Neunmalklugen
-- Siebengescheite machen. (Werner Mitsch, dt. Aphoristiker, 1936-)
-- Supercomputing in Magdeburg @ http://tina.nat.uni-magdeburg.de
--                 Heiko Bauke @ http://www.uni-magdeburg.de/bauke

Re: new testing version 1.7.90

[Brian Gough]

Thanks to everyone for all the testing. I am making a new 1.8 release now.

-- 
Brian Gough